No judgment. Anyone who tells you they’re not in the same boat is lying or on vacation…

The software companies whose whole pitch was that AI agents will replace your headcount watched their own stock get taken to the woodshed, because investors finally asked the same question you have been asking on every renewal call: if the agent does the work, why am I still paying per seat.

On June 29 the enterprise software index had its worst day of the year, and the same vendors are now inventing brand new fees to charge you when your agent touches the data you already pay them to store. Meanwhile, actual cybersecurity vendors got breached… through a sales battlecards tool. If you wrote this as fiction, your editor would send it back for being too on the nose.

It’s already starting off as a good week! More on this below…

If you’ve been reading, you know the deal by now… six stories, a contracting section with a fresh AI clause you can steal, a vendor fiscal calendar that tells you exactly whose sales rep is sweating this month, and every date worth putting on your calendar.

No fluff.

No corporate doubletalk.

Let’s go (but scream it like Tom Brady!).

The SaaS selloff is your renewal leverage. Use it before the vendors calm down.

If you renew any seat-based software this quarter, the single most useful thing that happened last week is that your vendor got scared. On June 29, enterprise software valuations cratered, with ServiceNow, Salesforce, Workday, and Adobe all down more than 35 percent on the year and ServiceNow trading at a forward earnings multiple of about 25 against a five year median near 145 (GuruFocus, June 29). The market is betting that agents erode seat revenue. That fear is your opening.

Here is the twist you need to watch on the paper. The same vendors are answering the seat erosion problem by building tollgates. ServiceNow rolled out Action Fabric, a layer that meters and charges you every time an outside AI agent touches data you already store in their platform, with pricing set per action (PYMNTS). Workday meters external agents per call through its Agent Gateway. SAP is routing agent traffic through its own hub. Translation for your budget: the thing you bought to save labor now carries a new usage line that did not exist in last year’s order form. If your redline does not address agent access, you are signing a blank check.

But at the same time, take a moment… enjoy this a bit. The disruptor got disrupted by its own slide deck.

Watch what they do about it, because that is the genuinely funny part. Having promised a world where agents roam free and do all the work, they are now sprinting to bolt a toll booth onto every door those agents walk through. They hyped the highway, cut the ribbon, let everyone on, then started charging by the mile. That is not the posture of a company calmly steering the future it invented. That is the posture of a company that lit the fire, realized it never sold fire insurance, and is now quietly trying to expense the fire truck to you. None of it makes them evil. It makes them nervous… and nervous vendors sign better deals. So enjoy the show for exactly one minute, then open the redline while they are still rattled.

DO THIS WEEK

Pull every seat based renewal closing before September and split the conversation in two. First, press for a real discount now while the vendor is defending its seat count, and ask for a multiyear price lock rather than a one year concession. Second, add one sentence to the order form: authorized AI agents acting on the customer’s behalf may access customer data at no incremental per action or per call charge for the term. If they will not agree, get the metered rate capped in writing before you sign, not at true up.

The FAR rewrite hit the Federal Register. Comments close July 23, and yes, commercial buyers should care.

Even if you never touch a federal contract, the clause library you quietly copied into your commercial templates is being renumbered under you. On June 23, the FAR Council published the first four proposed rules of the Revolutionary FAR Overhaul, covering FAR Parts 1, 2, 3, 4, 5, 6, 7, 10, 18, 24, 26, 29, 33, 37, 39, 40, 41, 49, 52, and 53, and the comment window is the bare minimum 30 days, closing July 23, 2026 (White House, June 25). If your contracting officer starts citing Part 40 for security requirements and your matrix still points at Part 39, that mismatch is now yours to explain.

Two changes matter for how you actually negotiate. Public award announcements are no longer required below 5.5 million dollars, up from 4.5 million, so competitive intelligence you used to pull from award notices gets thinner (GovExec, June 23). Separately, effective for defense contracts entered after June 30, the certified cost or pricing data threshold jumped to 10 million dollars, which reshapes what pricing support a defense supplier owes you (Government Contracts Law). This is a rare moment where the rulebook is genuinely open to comment. Silence is a choice.

DO THIS WEEK

Run a find and replace audit on your contract templates and compliance matrices for hard coded FAR and DFARS clause numbers, because new solicitations are already using the revised numbering. If any part of your business sells to or buys from the government, read the four proposed rules at regulations.gov and file a comment before July 23. A two paragraph comment from a real practitioner carries more weight than you think, and it is free.

USMCA did not get renewed, and a 10 percent surcharge quietly sunsets around July 24.

If you source anything from Canada or Mexico, two clocks are running and both land this month. On July 1, the United States declined to renew USMCA in its current form, which does not kill the agreement but drops it into an annual review process that can run through 2036 (CNBC, July 1). Qualifying goods stay duty free for now, so this is not a cliff. It is something worse for planning purposes: a prolonged stretch of not knowing the rules, with automotive rules of origin and Chinese transshipment squarely on the table.

The nearer term item is the money one. The 10 percent Section 122 surcharge on goods that do not qualify under USMCA is scheduled to expire around July 24, after which non qualifying goods could revert toward standard most favored nation rates of roughly 3 to 4 percent, while the 50 percent Section 232 tariffs on steel and aluminum stay in force regardless of origin (Steptoe). If you have been paying that surcharge because qualifying looked like too much paperwork, the math on getting compliant just changed, and the merchandise processing fee exemption is still sitting there for USMCA qualifying goods.

DO THIS WEEK

Pull your top 20 cross border SKUs by spend and confirm, in writing from the supplier, whether each one qualifies under USMCA rules of origin today. For anything non qualifying, ask your customs broker to model landed cost after the July 24 surcharge change before you cut the next PO. Then add a tariff and origin change trigger to any long term supply agreement you sign this quarter, so a rules of origin shift becomes a repricing conversation, not a silent margin hit.

RAMmageddon is now hitting your invoices. Memory jumped up to 89 percent in one quarter.

We’re probably going to talk about this a lot for a while.

Anyone who buys hardware, servers, or anything with memory in it needs to stop treating this as a tech press story. Consumer and commodity DRAM prices rose by as much as 89 percent quarter over quarter in Q2 2026, with a 16 gigabyte DDR4 module going from about 137 dollars to 207 dollars in a single quarter (Wccftech, June). This is not a normal cycle. Samsung, SK Hynix, and Micron are pouring wafer capacity into high bandwidth memory for AI data centers, and every HBM wafer is a wafer denied to the ordinary DRAM in your laptop fleet and server refresh.

The signals last week were loud. Micron’s CEO said the shortage runs through 2027 and only eases in 2028. Apple hiked prices on June 25 across iPads, Macs, and more, and the stock had its worst day in over a year on the news. Samsung, SK Hynix, and Micron are also now facing a price fixing class action tied to the shortage. Practically, that means OEMs like Dell, HP, and Lenovo are pushing 15 to 20 percent hardware increases and shortened quote validity windows. The buyer who waits for prices to normalize is making a bet that every analyst says will lose through at least 2027.

DO THIS WEEK

Pull forward any hardware refresh you can justify in the next two quarters and lock it now with your reseller, because early orders are effectively setting deployment schedules. Standardize on a short list of approved memory and server configurations so a shortage does not force a last minute redesign. On any purchase order, insist on firm pricing with a fixed quote validity date rather than price at time of shipment, and ask your reseller to reserve and hold inventory against your roadmap.

Your unlimited AI seat deal is gone. GitHub Copilot moved to usage pricing, and the bills are landing.

If your company is scaling AI agents, the budgeting model you built on per seat logic already broke, you just may not have seen the invoice yet. Per token prices fell roughly 98 percent since early 2024, and enterprise AI bills still went up an estimated 320 percent, because agentic workflows burn 5 to 30 times more tokens per task than a chatbot query (Henon, sourced to industry data). Uber burned through its entire annual AI budget by April. The share of FinOps practitioners responsible for managing AI spend went from 31 percent to 98 percent in a year (Optimum Partners). This is now a procurement problem, not an engineering curiosity.

The pricing model is moving under you in real time. GitHub Copilot shifted to usage based pricing credits on June 1, and the broader pattern is that roughly 85 percent of SaaS providers have moved to hybrid or consumption pricing tied to usage (elvex). The trap has a name now, token maxing, which is defaulting every task to the most expensive frontier model with no routing or governance. The fix is not buying less… it is contract terms that give you visibility and a ceiling before the always on agents run up a bill nobody approved.

DO THIS WEEK

For every AI or agent product on consumption pricing, get three things in the contract: a monthly spend cap or hard alerting at a threshold you set, committed use discount tiers you can grow into rather than commit to blind, and the right to route workloads to cheaper models without penalty. Ask finance for a monthly token variance report the same way you get a cloud dashboard. If a vendor cannot show you per team usage, that opacity is the product working as designed, and it is negotiable.

A sales tool got breached and took roughly 200 companies with it, including the security vendors.

This is the story that should make you open your vendor integration list today. Between June 11 and June 24, attackers compromised Klue, a market intelligence and battlecards SaaS tool, harvested OAuth tokens through a legacy credential, and pivoted into the connected Salesforce environments of downstream customers, hitting close to 200 organizations (Rescana). Reported victims included LastPass, Huntress, Recorded Future, Tanium, Jamf, Gong, and HackerOne (Bright Defense breach log). The security companies you buy from got hit through a fourth party nobody was watching.

The mechanism is the lesson. Nobody broke down Salesforce’s door. A trusted integration held an OAuth token, and that token was the door. Salesforce disabled the Klue integration on June 17, but by then the data had walked. Industry breach data now puts the blast radius at more than five downstream victims per vendor incident, and the average public disclosure lag near 117 days (Black Kite). Your annual security questionnaire did not catch this, because the risk was a live token, not a policy on paper.

DO THIS WEEK

Pull the list of third party apps with OAuth grants into your CRM and core systems this week, and revoke every integration nobody can name an owner for. For the ones you keep, confirm the vendor rotates and scopes credentials and does not carry legacy ones. Then check your contracts for a breach notification clock measured in hours, not the vague reasonable time that let this sit. Continuous monitoring beats the annual questionnaire, because a vendor clean in Q1 can be the entry point in Q2.

Standard disclaimer from the guy with the law degree who is not being your lawyer: this is what I have learned redlining these deals,... And this is not legal advice. Tie your review to what actually happened this week, because that is where the leverage and the risk are both hiding. Four tips, each anchored to a story above.

1. The AI clause to steal this week: data, training, and zero retention

Tied to the AI spend and agent tollgate stories. Any product that routes your data through a model needs an explicit answer on training. The market standard position in 2026 is opt out at minimum, opt in for any model training, with zero retention as the fallback when the vendor will not commit (GC AI). Language you can adapt:

“Vendor shall not use Customer Data to train, fine tune, or improve any model except to generate output for Customer. Vendor shall not retain Customer Data for any model training purpose. Any authorized AI agent acting on Customer’s behalf may access Customer Data without additional per action or per call fees during the term.”

That last sentence is the one that answers the Action Fabric and Agent Gateway metering. Put it in the order form, not just the linked terms, because linked terms change without your signature.

2. Cap indemnity to real insurance, not 12 months of fees

Vendors love to cap indemnifiable claims at the fees you paid, which on a small deal can be a rounding error against a real IP or data claim. Tie the liability cap for indemnification to the vendor’s actual errors and omissions coverage instead. If they carry 5 million dollars in coverage, the cap for indemnifiable claims should be 5 million, not 60 thousand in annual fees (Levy, Contracting with AI Vendors). This aligns the incentive: the vendor either keeps real coverage or accepts real exposure. Here is the 2026 gotcha, though, standard insurers have started adding generative AI exclusions to commercial policies, so a vendor’s indemnity can be genuinely unfunded (Honigman). Ask for a certificate of insurance and confirm AI related claims are not excluded before you rely on that indemnity.

3. Breach notification measured in hours, and OAuth hygiene as a covenant

Tied to the Klue breach. The clause that failed those 200 companies was almost certainly a vague notification standard. Specify a hard clock, notice within 48 or 72 hours of the vendor becoming aware, plus the right to disable integrations immediately on suspicion. For anything with an OAuth grant into your systems, add a covenant that the vendor rotates and scopes credentials, does not retain unused legacy credentials, and notifies you of any credential compromise affecting the integration. A questionnaire is a snapshot. A covenant is a promise you can enforce.

4. Calendar the auto renewal, because you lose leverage on day one without it

The most boring tip is the one that saves the most money. Standard auto renewal clauses lock you into a renewal at the existing price plus an escalator unless you give 60 to 90 days notice, and the teams that do not calendar that date walk into the renewal with zero leverage (GC AI). With the SaaS selloff making vendors nervous about seat counts, the notice window this year is not just an exit, it is your invitation to renegotiate price and terms while they are motivated. Build a renewal tracker with the notice date, not the renewal date, as the trigger.

Every major enterprise vendor whose fiscal year or fiscal quarter closes within the next 60 days, meaning through early September, with the current business context that moves your leverage and one thing to do this week. Quarter and year end is when the discount authority is highest and the rep is most motivated. Know the date and you set the tempo.

Palo Alto Networks Fiscal year ends July 31

Leverage read: This is the big one. Their entire fiscal year closes July 31, they just raised full year guidance after a standout Q3, and Nikesh Arora is pushing customers hard toward platformization, with CyberArk and Chronosphere freshly integrated (Q3 results, June 2). Reps have annual quota and platform count targets riding on your signature.

Do this week: If you have any PANW spend, your rep needs the deal before July 31. Bundle your renewals into a platformization conversation and trade a multiyear commitment for a real discount and a firm price lock, not a one time credit.

Cisco Fiscal year ends around July 25

Leverage read: Cisco closes its fiscal year on the last Saturday of July, roughly July 25. On top of the normal year end push, memory cost inflation is squeezing hardware margins, which cuts both ways: they want the booking, but list creep is real.

Do this week: Time any hardware or enterprise agreement renewal to the final two weeks of July. Ask for the year end incentive explicitly, and if Splunk is in the mix, make them cross sell for the discount rather than the other way around.

Salesforce Q2 fiscal year 2027 ends July 31

Leverage read: Salesforce stock has lagged peers on the fear that Agentforce cannibalizes seats, and Agentforce runs on consumption pricing at roughly 10 cents per action (pricing overview). A nervous vendor at quarter end is a flexible vendor.

Do this week: Push for flex credits you can grow into rather than a large committed Agentforce spend, and get a price protection cap on per action rates. Use the quarter close to lock any core CRM renewal at a real discount now.

Workday Q2 fiscal year 2027 ends July 31

Leverage read: Workday has been the most cautious major on AI messaging, but it is now metering external agents per call through Agent Gateway and bundling Sana as its front door. Quarter end lands July 31.

Do this week: Before your agent traffic scales, get the per call metering rate capped and confirm authorized agents can access your Workday data without a surprise usage line. Anchor renewals to the July 31 close.

Oracle Q1 fiscal year 2027 ends August 31

Leverage read: Oracle is riding high on AI cloud and remaining performance obligation growth, so they are less discount hungry on OCI. License and support renewals are a different animal and still very negotiable at quarter end.

Do this week: Separate the OCI enthusiasm from your on premise license and support renewal. Audit any unlimited license agreement position before August 31 so you are not certifying under time pressure at their quarter close.

Accenture Fiscal year ends August 31

Leverage read: Accenture’s full year closes August 31, and consulting is under visible margin pressure as generative AI deflates billable hours. Year end is when statement of work and rate card flexibility is highest.

Do this week: Reopen rate cards and open statements of work before August 31, and ask directly for a generative AI productivity credit, since they are using the tools to deliver your work faster.

Microsoft Fiscal year 2027 began July 1

Leverage read: Microsoft’s fiscal year ended June 30, so reps are now carrying fresh annual quota. Counterintuitively, the start of the year is a strong window: reps want to bank pipeline early, and their next quarter end is September 30.

Do this week: If you have an Enterprise Agreement or Copilot expansion in flight, engage now while reps are hungry to build the year’s pipeline, and hold your leverage for the September 30 quarter close if they stall.

Also closing July 31, worth a call if you hold spend:

● Zscaler, Nutanix, and Intuit all close their fiscal year on July 31. Full year quota, maximum discount authority.

● CrowdStrike, Snowflake, and Dell all close a fiscal quarter July 31. Quarter end incentives apply.

● AWS, SAP, IBM, and ServiceNow run on the calendar, so their next close is September 30, just past this window. Start positioning now so you are not rushing a Q3 deal in late September.

Two negotiation angles the news just handed you:

● The June 29 software selloff means seat based vendors are defending renewals. Trade a multiyear commitment for a genuine discount while the fear is fresh, and lock price before they stabilize.

● The memory shortage means hardware quotes have short shelf lives. Treat any hardware discount as perishable and get firm pricing with a validity date, not price at shipment.

Everything worth a calendar hold between now and early August, with why it matters to you specifically, not just what it is.

Now through July 24

The Section 122 10 percent surcharge on non USMCA qualifying goods is scheduled to sunset around July 24. If you import from Canada or Mexico, this is the window to confirm qualification and remodel landed cost before the change, not after.

July 23

Comments close on the first four Revolutionary FAR Overhaul proposed rules. This is your one clean shot to influence the biggest federal acquisition rewrite in four decades, and the comment is free to file at regulations.gov.

Around July 25

Cisco’s fiscal year ends. If you have Cisco hardware, enterprise agreements, or Splunk in play, this is peak discount authority. Have your ask ready before the last week.

July 31

The heavy date. Palo Alto Networks, Zscaler, Nutanix, and Intuit close their fiscal year, and Salesforce, Workday, CrowdStrike, and Snowflake close a quarter. If any of these renewals sit in your queue, the rep’s motivation peaks here. Do not let the date pass unused.

Dell closes its fiscal quarter, colliding with peak memory price pressure. If a server or fleet refresh is coming, this is a timing and pricing conversation, not a wait and see.

On the horizon:

August 31 Oracle’s fiscal quarter and Accenture’s fiscal year both close. Just outside 30 days, but ULA audits and consulting rate card resets take longer than a week to prepare, so start now.

Already in effect:

June 30 The certified cost or pricing data threshold for defense contracts rose to 10 million dollars for contracts entered after June 30. If you buy from defense suppliers, expect changes in what pricing support they will volunteer.

Ongoing:

The USMCA annual review process is live, with United States and Mexico bilateral talks continuing past the July 1 deadline. Treat North American sourcing as a prepare and decide situation for the rest of the year, not a wait and see one.

That’s a wrap. Now, go fix something for someone who failed to plan properly.

SOURCE INDEX

Primary sources first, cited inline above and listed here in full. Where a primary filing was summarized, the summarizing source is noted.

1. White House, OMB Advances Revolutionary FAR Overhaul (June 25, 2026) https://www.whitehouse.gov/releases/2026/06/omb-advances-revolutionary-far-overhaul-with-formal-publication-of-regulatory-changes/

2. Federal Register, FAR Overhaul Parts 1, 2, 4, 33, 39, 40, 53 (June 23, 2026) https://www.federalregister.gov/documents/2026/06/23/2026-12559/federal-acquisition-regulation-revolutionary-federal-acquisition-regulation-overhaul-parts-1-2-4-33

3. Federal Register, FAR Overhaul Parts 6, 7, 10, 18, 26, 37, 41 (June 23, 2026) https://www.federalregister.gov/documents/2026/06/23/2026-12560/federal-acquisition-regulation-revolutionary-federal-acquisition-regulation-overhaul-parts-6-7-10-18

4. SBA Office of Advocacy, Newly Released FAR Proposed Rules (June 25, 2026) https://advocacy.sba.gov/2026/06/25/newly-released-federal-acquisition-regulation-proposed-rules/

5. Government Executive, Federal acquisition overhaul moves to proposed rules (June 23, 2026) https://www.govexec.com/management/2026/06/federal-acquisition-overhaul-proposed-rules/414348/

6. Government Contracts Law, FY2026 NDAA CAS and Cost or Pricing Data Thresholds https://www.governmentcontractslaw.com/2026/01/swept-away-fy26-ndaa-updates-to-cas-and-certified-cost-or-pricing-data-thresholds/

7. CNBC, U.S. will not renew USMCA (July 1, 2026) https://www.cnbc.com/2026/07/01/trump-usmca-canada-mexico-trade-treaty.html

8. Steptoe, North American Trade at a Crossroads on USMCA Review https://www.steptoe.com/en/news-publications/stepwise-risk-outlook/north-american-trade-at-a-crossroads-on-the-deadline-of-usmca-review.html

9. Plante Moran, USMCA review 2026: what manufacturers should do now https://www.plantemoran.com/explore-our-thinking/insight/2026/07/usmca-review-2026

10. Wccftech, DRAM prices surge up to 89 percent in Q2 2026 https://wccftech.com/memory-shortages-destroy-consumer-segment-dram-prices-surge-89-percent/

11. Wikipedia, 2025 to present global memory supply shortage (Micron CEO, Apple) https://en.wikipedia.org/wiki/2025%E2%80%93present_global_memory_supply_shortage

12. The Register, Expect more DRAM price hikes (June 2, 2026) https://www.theregister.com/storage/2026/06/02/expect-more-of-those-dram-price-hikes-as-memory-shortage-continues-to-bite/5250049

13. GuruFocus, Software Giants Face Valuation Contraction (June 29, 2026) https://www.gurufocus.com/news/8936415/software-giants-face-valuation-contraction-amid-ai-concerns-now

14. PYMNTS, ServiceNow, SAP and Workday Make AI Agents Pay to Play https://www.pymnts.com/artificial-intelligence-2/2026/servicenow-sap-and-workday-make-ai-agents-pay-to-play/

15. Henon Insights, Token prices fell 98 percent, enterprise AI costs tripled https://henon.ai/insights/token-prices-fell-98-enterprise-ai-costs-tripled-now-companies-are-asking-why

16. Optimum Partners, AI Token Costs and how they might wreck your budget https://optimumpartners.com/insight/ai-token-costs-and-how-they-might-wreck-your-budget/

17. elvex, AI Token Cost Enterprise: Stop Budget Blowouts in 2026 https://www.elvex.com/blog/ai-token-cost-enterprise-budget-control

18. Rescana, Klue Supply Chain Breach (June 2026) https://www.rescana.com/post/klue-supply-chain-breach-exposes-oauth-tokens-and-salesforce-data-in-multi-stage-cybersecurity-incident-june-2026

19. Bright Defense, Recent Data Breaches log (Klue victims) https://www.brightdefense.com/resources/recent-data-breaches/

20. Black Kite, 2026 Third Party Breach Report https://blackkite.com/reports/third-party-breach-report-2026

21. Palo Alto Networks, Fiscal Q3 2026 results (June 2, 2026) https://www.paloaltonetworks.com/company/press/2026/palo-alto-networks-reports-fiscal-third-quarter-2026-financial-results

22. GC AI, A SaaS Agreement Has up to 50 Clauses. Six Carry the Risk. https://gc.ai/blog/saas-agreement

23. Levy, Contracting with AI Vendors (2026) https://storage.ghost.io/c/44/95/449506ca-034e-480f-9725-fcde08ef1cc1/content/files/2026/03/Contracting-with-AI-Vendors.pdf

24. Honigman, The AI Insurance Gap and What It Means for Technology Contracts https://www.honigman.com/the-matrix/ai-insurance-gap-what-it-means-for-technology-contracts

Procurement Unmedicated is written by Dan Shibilia. Forward it to the one

USMCA Review Starts Tomorrow. If You Source Anything From Canada or Mexico… This Is Your Problem.

Here is the plain version of what is happening. The US, Canada, and Mexico built a mandatory review into their trade agreement from the start. That review officially opens July 1st. The three countries now have to formally decide whether to extend USMCA for another 16 years or not. If they do not reach consensus, the agreement shifts to annual reviews until 2036. It does not expire tomorrow. But the uncertainty it creates is real, immediate, and expensive.

Why does this affect your week-to-week job? Because roughly $1.8 trillion in annual North American trade flows under USMCA. That is IT services from Canadian firms, marketing spend with Mexican agencies, packaging and components manufactured in both countries, staffing arrangements, logistics, chemicals, and practically every category of indirect spend that you have ever sourced from a North American supplier. USMCA compliance is what exempts those imports from the tariff stack. The 83.9% of Canadian and Mexican imports currently claiming USMCA exemption are doing so because it is the difference between paying zero additional tariffs and paying 10 to 35% on top of base rates.

A clean 16-year extension is unlikely. CSIS rates it as the least probable outcome. The more realistic scenario is a prolonged negotiation that stretches into late 2026 or 2027 while the core agreement stays in force. That sounds like a win, except that protracted negotiations mean rules of origin could tighten mid-negotiation, sector-specific concessions could disrupt specific supply lanes, and investment decisions that depend on North American sourcing stability get deferred. Your CFO is not going to green-light a 10-year supplier commitment while the trade framework governing that commitment is in active renegotiation.

Source: Convera: USMCA 2026 Review Analysis | CSIS: Six Scenarios for USMCA’s Future | Tax Foundation: USMCA Tariff Impact | BSI: USMCA Review What You Need to Know

The Government Just Wrote AI Governance Into Contracts. Commercial Contracts Are Next. Here Is What It Means.

On June 17th, GSA published a proposed contract clause that will eventually land in every federal contract where an AI tool touches government data. GSAR clause 552.239-7001 requires contractors to disclose all AI systems being used to perform work within 120 days of starting a contract, report any data incident within 72 hours, and prohibit using government data to train or improve AI models without explicit authorization. It also requires prime contractors to flow those obligations down to their subcontractors and AI vendors.

I know what you are thinking. Government contracts are their own world. What does this have to do with me? Two things.

First, if your company sells anything to the federal government, or if any of your indirect suppliers do, this clause will eventually land in your contracts. The GSA schedule covers practically every professional services category. IT services, consulting, staffing, facilities management, and marketing are all on the schedule. When this clause gets finalized and applied to GSA vehicles, it flows to every contractor on those vehicles, including the ones you are buying from today.

Second, and this is the bigger story: commercial contracting follows government contracting. It’s just how things work. Liability standards, data protection requirements, cybersecurity clauses, and performance obligations that started in federal acquisition have found their way into commercial SaaS agreements, professional services contracts, and outsourcing agreements over the past two decades. The government is now writing the first detailed, enforceable AI governance language in contracts. Your procurement and legal teams should be reading this clause today, because it is a preview of what your enterprise clients are going to start demanding in your own contracts within the next 12 to 24 months.

Source: Federal Register: GSAR LLM Safeguarding Clause, June 17, 2026 | Singularity Capital Advisors: GSA AI Rule, June 2026 | Mondaq: GSA LLM Rewrite Analysis, June 23, 2026

That $800 Duty-Free Threshold You Forgot About? The Government Just Made Sure You Cannot Get It Back.

The de minimis exemption is the rule that let packages valued under $800 enter the US without paying duties. Vendors used it heavily. Procurement teams relied on it for sample shipments, low-value MRO, lab supplies, and tail spend. The current administration suspended it. Courts pushed back, ruling that the IEEPA authority used to suspend it was unconstitutional.

Here is what happened on June 23rd: CBP published two new proposed rules using a completely different legal authority, specifically the Tariff Act of 1930, to justify keeping de minimis suspended permanently. The IEEPA-based suspension was vulnerable to the same legal challenge that killed the broader tariff regime. The Tariff Act of 1930 is not. CBP is basically inoculating the de minimis suspension against the court challenge that was most likely to work.

Here is what this means in plain English. If your sourcing model assumed the de minimis exemption would come back when the courts sorted things out, that assumption needs to be retired. Penn Wharton’s June 16th tariff rate update confirms the broader tariff landscape has settled at an average effective rate of 7% as of April. That is the floor. Plan from it.

Source: RVIA: Latest Tariff Developments, June 25, 2026 | Penn Wharton: Effective Tariff Rates June 16, 2026

70% of Organizations Have Nobody In Charge of their SaaS Spend. Is One of Them Yours?

Let us start with the number that should be uncomfortable. Zylo’s 2026 SaaS Management Index found that only 31% of organizations have clearly defined ownership between FinOps, IT, and procurement for SaaS spend. That means at roughly 7 out of 10 companies, when something goes wrong with software costs, three departments all point at each other. Sound familiar to you, it does to me…

The average organization manages 305 SaaS applications and spends $55.7 million annually. Total SaaS spend rose 8% year over year. AI-native app spending inside large enterprises is up 393%. The average company adds nine new applications every month. And 46% of the applications in the average portfolio carry a poor or low security risk score, although most companies aren’t risk scoring at all. That last number is not a security team problem. It is a procurement problem. Every one of those applications was purchased by someone using a company card or a purchase order. Somebody approved that spend.

One specific development worth noting: Salesforce announced on June 15th that it is acquiring Fin, formerly known as Intercom, for $3.6 billion. Fin resolves an average of 76% of customer support volume end-to-end without human escalation. This is being folded into Agentforce. What does that mean for procurement? The next time your Salesforce account executive sits down with you, they are going to have a fully autonomous customer service agent to sell you. That agent runs on per-action consumption pricing. Given what we have all watched happen with AI token costs this year, that is a model you want to pilot with a spending cap, not commit to at scale.

Source: Zylo: 2026 SaaS Statistics and Trends | Tropic: 2026 SaaS Procurement Predictions | TechnologyChecker: Salesforce Acquires Fin, June 15, 2026

DRAM Shortages Through 2030. PCB Lead Times Up. This Is Not a Cycle. It Is the New Baseline.

For everyone on the direct sourcing and supply chain side, here is the honest state of play. The Hormuz reopening helped freight. It did not fix materials.

Supplyframe’s supply chain intelligence published this week identified two structural problems that will not resolve this year or next. First, printed circuit board supply shortages created by the Iran conflict are persistent. Lead times are elevated through 2026 with only modest improvement projected in early 2027. Pricing is inflexible. If you have any electronics manufacturing, industrial controls, or hardware procurement in your category portfolio, your Q4 and Q1 numbers are going to reflect this whether your contracts acknowledge it or not.

Second, and this is the one that gets underreported: AI infrastructure is competing with your direct materials for the same components. DRAM, specialized resins, rare earth materials, and advanced semiconductors are all being consumed by data center buildout at a pace the supply chain was not designed to support. Industry experts are projecting the DRAM shortage lasts until 2030. Not 2026. Not 2027. 2030. If you are running just-in-time inventory models on components that sit in that supply chain, you need to have a frank conversation with your operations team about safety stock and dual sourcing.

Source: Supplyframe: 3 Biggest Supply Chain Risks in 2026 | Thomson Reuters: 2026 Supply Chain Challenge Report | Deloitte: 2026 Manufacturing Industry Outlook

The Contract Review Section Nobody Gave You When You Started. You Are Welcome.

I am a attorney by training and a procurement guy by accident. I have reviewed and negotiated more contracts than I can count across technology, professional services, logistics, staffing, and manufacturing and the single most consistent problem I see across procurement organizations is not that people do not care about their contracts. It is that nobody gave them a framework for what to actually look for or, in some cases, if they did the framework is from 2001.

So here it is. Not a legal treatise. A practitioner’s guide. The questions to ask, the things to look for, and right now especially, what to do about AI language in contracts.

(I’m working on a full Contract Review Framework in my “free time.”)

BEFORE YOU SIGN ANYTHING: THE QUESTIONS MOST PEOPLE SKIP

Most contract review processes start at redlining clauses. They should start at understanding the actual commercial deal… I’ve talked about this before online… go read it. Before you open a contract to review, you should be able to answer these questions clearly:

1. What exactly are we buying and how is it measured? If you cannot define the deliverable in one sentence, the contract probably cannot either. Vague scope is how you end up in a dispute 18 months later about whether something was in scope.

2. What is the exit? How do you get out of this agreement if the vendor underperforms, gets acquired, raises prices dramatically, or you simply do not need the service anymore? Termination for convenience, notice periods, and automatic renewal clauses are where procurement value gets silently destroyed.

3. Who owns what was created? If a vendor is building something for you, developing code, creating content, designing a process, or producing deliverables of any kind, IP ownership needs to be explicit. Default contract law varies by state and by what was actually paid for. Do not assume you own it because you paid for it.

4. Where does liability actually sit? Most standard vendor agreements cap the vendor’s liability at one year of fees paid. On a $10 million annual contract, that is $10 million. On a $500K annual contract where a data breach causes $50 million in downstream damage, that cap is not going to cover you. Know what you are signing.

5. What data are you handing over and what can they do with it? Every SaaS agreement, every managed services contract, every staffing arrangement involves data moving somewhere. Understand what data leaves your environment, where it goes, who can access it, and whether the vendor can use it for anything beyond performing your contract.

THE AI LANGUAGE SECTION: WHAT TO ASK FOR RIGHT NOW

This is the most time-sensitive part of this section. Your existing vendor contracts almost certainly say nothing about AI. They were drafted before AI tools were a standard part of service delivery. That silence is now a liability.

Here is the scenario that keeps us attorneys up at night. Your vendor is using an AI tool to deliver your contract. The AI makes an error. The error causes you financial harm. You go to enforce your contract. Your contract says nothing about AI. The vendor argues it just used commercially reasonable tools to perform the work. You argue the AI output constitutes a deliverable under the contract. The indemnification clause was written for software bugs, not autonomous decision-making. Nothing about this is quick or cheap.

The four AI questions every contract negotiation needs right now:

1. Is the vendor using AI tools to perform any part of this contract? If yes, which ones? This should be a mandatory disclosure, not a voluntary one. Require a schedule of AI systems used in contract performance and require written notice before any changes to that schedule.

2. What data will be processed by those AI tools? Specifically: will any data you provide, including prompts, inputs, outputs, or metadata, be used to train, fine-tune, or improve the vendor’s AI models? The answer should be no. Get that in writing. Many vendor AI tools have model improvement clauses buried in their terms of service. Your contract should supersede those.

3. Who is liable when the AI makes a decision that causes harm? The vendor’s position will be that AI outputs are tools, not deliverables, and liability sits with the human who acted on them. Your position should be that the vendor is responsible for the quality and accuracy of AI-assisted work product delivered under the contract. These are reconcilable positions. But you have to have the conversation during formation, not after a problem.

4. What happens when the AI tool changes? Model updates, vendor changes, architecture shifts, and capability degradations can all change the quality of AI-assisted deliverables without triggering a contract modification. Require notification of material changes to AI systems used in contract performance and include a right to re-evaluate SLAs if those changes affect output quality.

One more thing worth saying. The GSA clause we covered in Story 2 defines government data broadly to include user prompts and query outputs. That is more expansive than most commercial contracts contemplate. Before you dismiss it as a government contracts issue, ask yourself: how would you feel if your most sensitive business data ended up in a vendor’s training set? That is the risk the clause is designed to prevent. It is not a bad model to adapt for your own vendor agreements.

Your IT Hardware Budget Was Built for a Memory Market That No Longer Exists.

Gartner’s research director said it plainly: “The speed at which the memory pricing has increased has shocked everybody.” That is not a vendor trying to upsell you. That is an analyst describing what is happening to every IT hardware budget on the planet right now.

Here is the situation in plain terms. The three companies that control 95% of global DRAM production, Samsung, SK Hynix, and Micron, have made a rational business decision to redirect their manufacturing capacity toward high-bandwidth memory for AI infrastructure. HBM is more profitable. AI customers pay more and commit earlier. IDC confirmed it: every wafer allocated to an HBM stack for an AI GPU is a wafer denied to the conventional DRAM used in your enterprise servers, laptops, and storage systems. This is a zero-sum game and right now the AI buildout is winning.

The numbers are not subtle. TrendForce reported conventional DRAM contract prices rose 90 to 95% quarter over quarter in Q1 2026. The largest quarterly increase on record. Server DRAM is up more than 60% quarter over quarter as cloud providers lock in capacity through long-term agreements. HP reported in its Q1 earnings call that memory now accounts for 35% of the cost to build a PC, up from 15 to 18% previously. One component. 35% of the build cost.

Here is the number that should make you stop scrolling. DDR4 spot prices have now hit $2.10 per gigabit, which exceeds advanced HBM3e at $1.70 per gigabit. The old commodity memory costs more than the cutting-edge AI memory. When the legacy product is more expensive than the premium product, you are not in a normal supply cycle. You are in a structurally distorted market and your budget assumptions are probably wrong.

I know… it’s a lot of numbers that really mean much other than costs went up but there’s more…

WHY THIS HITS IT SOURCING SPECIFICALLY RIGHT NOW

Two problems are compounding at the same time and they are landing on the same desk.

Problem one: the Windows 10 end-of-life in October 2025 triggered enterprise refresh cycles that are now running directly into the shortage. Your organization may have already committed to a PC refresh wave this year with a budget built on Q3 2025 hardware pricing. Lenovo, Dell, HP, Acer, and ASUS have all warned of 15 to 20% PC price increases for 2026, with some reports noting increases as high as 30% as the shortage deepens. The refresh budget your team approved six months ago is likely already underwater. This is not a small variance. It is a material budget gap that needs to be in front of your finance team now, not at year-end review.

Problem two: lead times for larger DRAM orders have extended beyond 40 weeks. Not four weeks. Forty weeks. That makes many standard configurations unworkable for fiscal year 2026 planning. If you have a data center expansion, server refresh, or infrastructure project planned for Q3 or Q4 of this year, the memory you need to complete it may simply not be available on your timeline regardless of what you are willing to pay. That is not a price problem. That is an availability problem, and price alone cannot solve it.

STRUCTURAL, NOT CYCLICAL. THE DISTINCTION MATTERS.

Every time there is a supply crunch, the instinct is to wait it out. That instinct is wrong this time. IDC is explicit: this is not a cyclical shortage driven by a temporary demand spike or a logistics disruption. It is a potentially permanent strategic reallocation of global silicon wafer capacity. The manufacturers made a deliberate choice. They are not going to un-make it until AI infrastructure demand moderates or new fab capacity comes online.

New fab capacity does not arrive until 2027 at the earliest. Micron’s CEO has stated supply tightness will continue into 2027. Samsung’s next major HBM fab does not reach volume production until 2028. The gap between what the market needs and what it can produce is not closing this year. Organizations that are waiting for normalization to make hardware decisions are going to be waiting for a long time while paying elevated prices the entire time.

One more wrinkle worth knowing. Gartner projects a 130% year-over-year DRAM price rise for 2026 overall. IDC projects the PC market will shrink 11.3% as a result. That contraction means OEMs have less volume to spread fixed costs across, which puts additional upward pressure on the per-unit price you pay. The shortage is creating a reinforcing cycle and the enterprise IT buyer is sitting in the middle of it.

For all your parents or parents of gamers out there, this is why certain video games are no longer being produced in physical formats (CDs or whatever you call those cards that goes in the Switch.)

Source: Insight: 2026 RAM Shortage Enterprise Guide | SHI: What Is Driving the Global RAM Shortage | IDC: Global Memory Shortage Crisis Market Analysis | SoftwareSeni: DRAM Prices in 2026 Have Doubled | SHI: Impact of Memory Shortage on Data Center Buyers

Your Vendors Are On a Clock. Make Sure You Are On It Too.

Vendor fiscal year pressure is one of the few structural negotiating levers in procurement that is entirely predictable. Sales teams chasing quota close faster and more creatively than sales teams who just made their number. Here is the accurate calendar through end of September, with the vendors you are most likely negotiating.

CLOSING TODAY: MICROSOFT | FISCAL YEAR END JUNE 30

Today is the last day of Microsoft’s fiscal year. The M365 price increase goes live tomorrow, July 1st. These two things together, fiscal year close and price lock deadline, create the highest-pressure negotiating window in Microsoft’s calendar. If you have not already started a conversation, it is not too late to call your account executive today. Not email. Call.

What to do: If you have any Microsoft renewal in the next 12 months, ask today about accelerating it to lock in current pricing. Ask about bundling Unified Support into your EA separately. Ask for a license utilization report before you commit to quantities. The AI add-ons are expanding across M365 suites automatically starting this month. Do not renew at last year’s seat count without auditing actual usage first.

Source: SAMexpert: Microsoft AI Costs, Licensing vs Azure | AppDirect: Microsoft 365 Price Change Advisory, June 2026

JULY 31 | PALO ALTO NETWORKS FISCAL YEAR END

Q3 results released June 2nd showed 31% revenue growth, driven partly by the CyberArk acquisition. Their platformization strategy, consolidating customers onto fewer security products, means account teams are rewarded for expanding existing relationships in Q4. The urgency is real. Their Q4 is right now through July 31st.

What to do: If you have any Palo Alto Networks renewal in the next 12 months, start the conversation before July 31st. Their Q4 close combined with CyberArk integration complexity creates room for consolidation deals that bundle Prisma, Cortex, and CyberArk capabilities at better economics than you would see in August. Come in with documented competitive alternatives from CrowdStrike or Cisco. A verbal threat does not move Palo Alto. A documented quote does.

Source: Palo Alto Networks Q3 FY2026 Earnings, June 2, 2026

APPROXIMATELY JULY 26 | CISCO FISCAL YEAR END

Cisco Q3 earnings confirmed they are tracking toward their strongest full year ever with FY2026 guidance raised to $62.8 to $63 billion in revenue. A confident company, but Q4 quota pressure is still Q4 quota pressure. Cisco’s fiscal year closes on the last Saturday of July, approximately July 26th. Their account teams are in maximum push mode right now.

What to do: Cisco’s primary vulnerability right now is competition from cloud-native networking and the Microsoft-CrowdStrike partnership eating into their security business. If you have campus networking, Webex, or security renewals coming, document a competitive quote from a credible alternative before you negotiate. Cisco responds to documented alternatives far more than to theoretical ones.

Source: Cisco Q3 FY2026 Earnings Report, May 2026

JULY 31 | SALESFORCE Q2 CLOSE (FISCAL YEAR ENDS JANUARY 31)

Salesforce’s full fiscal year ends January 31st, but July 31st closes their Q2, which is a real pressure point. Their account teams are being pushed to hit mid-year targets. The Fin acquisition at $3.6 billion means they are actively trying to validate Agentforce’s commercial value. That creates a window where they need your commitment more than usual.

What to do: Use the Fin acquisition as leverage in any current Salesforce negotiation. Ask for an extended Agentforce pilot at no cost as part of your renewal terms. Commit to Agentforce consumption pricing only after you have real usage data, not projected usage data. The companies burning through AI budgets right now are the ones who committed to consumption volume before they knew what they were actually going to consume.

Source: Everest Group: Nine Tactics for Salesforce Negotiation | TechnologyChecker: Salesforce Fin Acquisition, June 15, 2026

AUGUST 31 | ACCENTURE FISCAL YEAR END

Accenture closes August 31st. Their commercial account teams are in Q4 right now. And they have a specific motivation this year beyond normal quota pressure: DOGE-related federal contract cancellations hit Accenture Federal Services directly. GSA placed them on the review list. Federal revenue pressure means their commercial teams are being pushed harder than usual to make up the gap.

What to do: If you have consulting, managed services, or IT outsourcing work with Accenture, their federal exposure is your negotiating context. They need to replace revenue. Use July and August to renegotiate rate cards, introduce performance-based pricing into statement-of-work language, or exchange term extensions for pricing concessions. Come with market benchmarks. Accenture has their own benchmarks and will use them against you if you come without data.

Source: Accenture Investor FAQ: Fiscal Year End August 31

Coming Up. All Explained. No Filler.

TODAY | JUNE 30 FAR Overhaul Phase 1 Live

Fixed-price contracting is now the federal default. If you have federal contracts or suppliers who do, the compliance clock is running.

JULY 1 USMCA Joint Review Opens

The first mandatory review of the US-Mexico-Canada trade agreement begins today. A clean extension is unlikely. Map your North American sourcing exposure before the negotiation creates uncertainty about your specific categories.

JULY 3 GSA AI Listening Session Registration Closes

If you want to present at GSA’s July 14th session on the LLM safeguarding clause, you must register by today. Virtual attendance is also available.

JULY 6 Section 301 Forced Labor Tariff Comments Close

The last day to submit written comments on the proposed 10% and 12.5% tariffs on 60 countries. If your supply chain touches any of those economies, this is your last formal opportunity to influence the outcome.

JULY 14 GSA AI Listening Session | GW Law School + Virtual

GSA’s public session on GSAR clause 552.239-7001. The AI governance language being debated here will eventually land in commercial contracts. Worth following even if you have no federal exposure.

JULY 26-31 Cisco + Palo Alto Networks + Salesforce Q2 Close Windows

Multiple fiscal year-end events in one week. Have your proposals and competitive documentation ready before this window, not during it.

JULY 29 Agency Deadline: Renegotiate Top 10 Federal Non-Fixed-Price Contracts

Every federal agency must have begun renegotiating its largest cost-reimbursement contracts to fixed-price terms. If you are a supplier on any of those, the call is coming.

AUGUST 3 GSA LLM Safeguarding Clause Comments Due

Last day for written comments. If your organization delivers services using AI tools and has any federal contract exposure, this is the last opportunity to formally shape how this clause develops.

AUGUST 31 Accenture Fiscal Year End

Maximum commercial pressure on Accenture account teams. If you have consulting or managed services work with them, the first two weeks of August are your negotiating window.

That is two weeks. Plus a contracting guide.

Seven stories, a full contracting section with actual questions to ask, and a vendor calendar through August. The USMCA story is the one that most indirect teams will underestimate because it feels like someone else’s problem until it suddenly is not. The AI contract language section exists because I have spent enough years in contract negotiations to know that the people who get burned are almost always the ones who assumed the other side was going to bring it up first.

They are not going to bring it up. You have to.

Drop a comment. Push back. Tell me what I got wrong. That is the whole point of this thing. I want to push a newsletter worth reading… something that helps you be better at your job. I want this to be a tool that makes you sound extra smart in that meeting. Help me build that.

SOURCES

1. Convera: USMCA 2026 Review Analysis

2. CSIS: USMCA Review 2026 Six Scenarios

3. Tax Foundation: USMCA Tariff Trade Agreement

4. BSI: USMCA Review What You Need to Know

5. Federal Register: GSAR LLM Safeguarding Clause, June 17, 2026

6. Singularity Capital Advisors: GSA AI Rule, June 2026

7. Mondaq: GSA LLM Rewrite, June 23, 2026

8. RVIA: Latest Tariff Developments, June 25, 2026

9. Penn Wharton: Effective Tariff Rates, June 16, 2026

10. Zylo: 2026 SaaS Statistics and Trends

11. Tropic: 2026 SaaS Procurement Predictions

12. TechnologyChecker: Salesforce Acquires Fin, June 15, 2026

13. Supplyframe: Supply Chain Risks 2026

14. Thomson Reuters: 2026 Supply Chain Challenge Report

15. Deloitte: 2026 Manufacturing Industry Outlook

16. Insight: 2026 RAM Shortage Enterprise Guide

17. SHI: What Is Driving the Global RAM Shortage

18. IDC: Global Memory Shortage Crisis Market Analysis

19. SoftwareSeni: DRAM Prices in 2026 Have Doubled

20. SHI: Impact of 2026 Memory Shortage on Data Center Buyers

21. Palo Alto Networks Q3 FY2026 Earnings, June 2, 2026

22. Cisco Q3 FY2026 Earnings, May 2026

23. Accenture Investor FAQ: Fiscal Year End August 31

24. SAMexpert: Microsoft AI Costs, Licensing vs Azure

25. AppDirect: Microsoft 365 Price Change Advisory, June 2026

26. Everest Group: Nine Tactics for Salesforce Negotiation

Last week was the first Monday Brief. This week we have fresh stories across all the categories that actually matter to people doing this work. No repeats, no filler. The Hormuz situation just broke wide open overnight. CLM is having an identity crisis. Your AI spend is already out of control and your CFO has not caught up yet. And if you have Microsoft, AWS, Oracle, or Salesforce renewals coming, there are specific moves you should be making right now.

Let us get into it.

The next seven days are not a normal week…. I know, I know… there are no normal weeks for us but this is going to be a bit different.

There are three categories of things happening simultaneously: hard regulatory deadlines with no extensions, vendor fiscal year pressure you can exploit right now, and market signals worth paying attention to. Here is each one and what it actually means for the work you do.

JUNE 22 | 11:59 PM EDT USTR Forced Labor Tariff Hearing Request Deadline

This is the use-it-or-lose-it deadline. June 22nd is the last day to formally request to appear before the USTR Section 301 Committee at the public hearing on proposed forced labor tariffs covering 60 countries including China, India, Vietnam, Canada, Mexico, the UK, and the EU. If your organization sources from any of these economies and you want to make a formal argument for product exclusions or a reduced tariff rate, you must file under docket USTR-2026-0266 by 11:59 PM EDT tonight. Miss this deadline and you are limited to written comments only. The proposed tariffs are 10% for 15 trading partners and 12.5% for the remaining 45. They are not in effect yet, but this is the moment to influence the final outcome. If you have not already spoken to your trade counsel about this, today is the day.

JUNE 30 | MICROSOFT PRICE LOCK Last Day to Lock In Pre-Increase M365 Pricing

Microsoft 365 prices increase July 1st. If your organization has M365 licenses renewing anytime in the next 12 months, renewing before June 30th locks in current rates for another year. This is not a rumor. Microsoft has confirmed the change. The window closes in 15 days. If you have not initiated this conversation with your account team yet, you have one business week to get it done. Frontline worker licenses are seeing the steepest percentage increases. Larger M365 deployments will see the biggest dollar impact. This is straightforward math: renewal now at current price versus renewal later at a higher price. There is no procurement strategy here. It is just a deadline.

JUNE 30 | MICROSOFT FISCAL YEAR CLOSE Maximum Quota Pressure on Microsoft Account Teams

June 30th is also the close of Microsoft’s fiscal year, which means every Microsoft account executive is in their highest-pressure 15 days of the year. This creates a real negotiating window that exists independently of the M365 price increase. Microsoft sales teams have end-of-year targets to hit. Buyers who are willing to close a deal before June 30th have leverage that simply will not exist on July 1st. This applies to Azure, Copilot, Dynamics, and the EA broadly, not just M365. If you have any Microsoft negotiation in progress or any renewal within the next six months, initiating or accelerating that conversation this week is not optional. It is the right procurement move.

JUNE 30 | FAR OVERHAUL PHASE 1 Federal Contracting Rule Changes Land

Phase 1 of the Revolutionary FAR Overhaul drops June 30th. The biggest change already in motion is the Executive Order directing all federal agencies to default to fixed-price contracts. OMB was required to issue implementation guidance to agencies by June 14th. The agencies themselves have until July 29th to review their 10 largest non-fixed-price contracts and begin renegotiating them to fixed-price or performance-based terms. If you are a supplier to the federal government, or a supplier to a supplier, this date starts the clock on renegotiation conversations you did not budget for. If you have time and materials contracts or cost-plus arrangements with federal agencies, now is the moment to understand your rights before a contracting officer comes to you with a modification request.

JUNE 30 | GENERAL MILLS FISCAL YEAR CLOSE A Large Buyer Finishing Their Year

General Mills closes its fiscal year on June 30th. This matters for anyone who supplies to them. Large companies in cost-control mode at fiscal year end slow approvals, tighten spend review, and sometimes push new purchase decisions to the next fiscal year. If you have open quotes, pending contract renewals, or new business discussions with General Mills, expect their procurement team to be heads-down right now. Timing your outreach for early July, when their new fiscal year begins with fresh budgets and renewed appetite to commit, is smarter than trying to close something in the next two weeks.

ALREADY IN EFFECT | JUNE 8 Section 232 Tariff Reductions on HVAC, Agricultural, and Industrial Equipment

This one already happened and most procurement teams have not caught up with it. On June 8th, a White House proclamation reduced Section 232 tariffs from 25% to 15% on agricultural equipment, certain residential HVAC systems and components, and mobile industrial equipment including bulldozers and forklifts imported from select countries. These reductions run through December 31, 2027. The category list is specific and requires HTS code-level review to confirm applicability. If you have any indirect spend in these categories, pull your customs entries from June 8th forward and verify you are being assessed at the correct rate. Overpayment on tariffs is recoverable but requires prompt action. If you are not sure whether your products qualify, your customs broker can confirm against the updated Annex III and Annex I-C product lists.

There are a few vendors worth calling attention to that will close their books for the year on June 30 and that matters if you are looking to renew a contract.

• Microsoft

• General Mills

• Micron Technology

Despite a post I saw on LinkedIn the other day, Oracle closed May 31st. Their Q4 is already over.

Nike closed May 31st.

FedEx just changed its fiscal year-end to December 31st, effective this month.

Accenture closes August 31st.

Salesforce closes January 31st.

AWS closes December 31st.

The reason this matters: vendor fiscal year timing is one of the few structural negotiating levers that is entirely predictable. But it only works if you know the actual date. Applying Q4 urgency pressure to a vendor whose Q4 ended three weeks ago is not a strategy. It is a waste of a conversation.

The vendors with real urgency on the table right now are Microsoft, because their fiscal year and their M365 price increase both land June 30th. Use that window. For everyone else, do the calendar math before you walk into the room.

Source: Microsoft Q3 FY2026 Earnings Call: Fiscal Year Guidance | Oracle FY2026 Annual Results, June 10, 2026 | Accenture Investor FAQ: Fiscal Year End August 31

The Strait of Hormuz Just Reopened. Your Contracts Are Still Broken.

This one landed yesterday and it is big. On June 14th, President Trump announced a peace deal with Iran is “now complete” and ordered an immediate end to the US naval blockade of Iranian ports in exchange for toll-free shipping through the Strait of Hormuz. Iran’s deputy Foreign Minister confirmed the agreement on Sunday. The G7 is discussing long-term implementation in Evian, France today.

To understand why this matters for us procurement professionals specifically, you need to understand what the last three and a half months actually did to your supply base.

The Strait has been effectively closed since the US and Israel launched strikes on Iran on February 28th. It carries roughly 20% of global oil trade. When it closed, crude oil and fertilizer prices spiked sharply, maritime insurance premiums went through the roof, hundreds of tankers were stranded or rerouted around the Cape of Good Hope, and energy feedstock costs hit plastics, chemicals, and packaging up and down the indirect supply chain. So, frankly… it’s caused a landslide of chaos that hit just about everything we touch.

Here is the procurement reality even now that a deal is struck: your contracts did not fix themselves overnight. If you have force majeure claims in flight, they do not automatically unwind. If your supplier repriced based on energy cost escalation, that new price is now your baseline unless your contract has a de-escalation mechanism. If you extended lead times and changed supplier commitments to compensate, those changes need to be formally addressed.

The reopening of the Strait is good news for markets and your 401k . It is not a reason to stop the contract housekeeping that the last three months forced onto your desk. If anything, it is the moment to lock in favorable terms before the urgency fades and your counterparties remember they have leverage again.

The G7’s stated goal is a “lasting reopening” with full freedom of navigation. Macron confirmed that today’s summit will address long-term Hormuz governance. That is a 60 to 90 day negotiation window at minimum. Plan accordingly.

Source: Al Jazeera: US-Iran Ceasefire Deal, June 14, 2026 | Axios: US-Iran Framework Agreement, June 14, 2026 | CSIS: The Strait of Hormuz in 8 Charts, June 11, 2026

You Do Not Know How to Contract for an AI That Acts on Your Behalf. Nobody Does.

For two decades, software contracts were about access. You paid per seat. You got uptime guarantees and a liability cap. The software did not do anything unless a human touched it. That model made legal risk manageable because humans were still the ones pulling the trigger.

Agentic AI just blew that model up. PYMNTS put the problem clearly: these systems approve refunds, reconcile invoices, negotiate with suppliers, trigger payments, and move data across systems. They do not wait for a human but your existing SaaS contract language was written for software that waited.

In May, LinkSquares launched what it calls the first all-agentic CLM platform, designed to draft, redline, and manage contracts autonomously… which is scary on so many levels. Microsoft’s Copilot Legal Agent for Word is doing clause-by-clause contract review. CLOC 2026 made clear that the features CLMs have charged premium pricing for are being absorbed by horizontal AI platforms that already live inside your document workflows.

If you are buying an AI tool that takes autonomous action inside your procurement or contracting process, your standard SaaS agreement is probably the wrong contract. You need indemnification language that accounts for autonomous decisions, clear definitions of what the system is authorized to do without human approval, audit trail requirements, and liability terms that reflect who bears the risk when the agent makes a bad call. Your legal team and your CLM vendor are almost certainly not aligned on this yet.

The practical move is to build an internal list of every AI tool in your stack that takes autonomous action… not just your CLM. Your AP automation. Your sourcing bots. Your intake platforms. Then ask one question about each: when this system acts without a human approving the specific action, who is liable for the outcome? If you cannot answer that from your current contract, you have work to do.

Source: PYMNTS: Contracting for Agentic AI, February 2026 | LinkSquares PR Newswire: All-Agentic CLM Launch, May 5, 2026 | Pramata: CLOC 2026 Recap, June 2026 | PwC: Agentic AI in Procurement, June 2026

You Are Losing 11% of Contract Value After Signature. Stop Pretending Otherwise.

The 2026 Contracting Benchmark Report, cited by Ironclad, puts a number on something procurement and contracting teams know in their gut but rarely quantify: organizations lose an average of 11% of contract value after signature through missed obligations, revenue leakage, and costs nobody is watching.

That number is not a technology problem. It is a process and attention problem. Most procurement teams negotiate hard to get a contract signed and then treat it like a filing event. Meanwhile, the SLA credits your vendor owes you go unclaimed. The volume discounts you negotiated only apply if you actually consolidate volume. The benchmarking rights that let you challenge pricing every 18 months sit unused because nobody remembered they existed.

The highest ROI thing most contract management teams can do right now costs nothing. Pull your top 20 contracts by spend value. Read the post-signature obligations section of each one. List every right, credit, or discount that requires your team to take affirmative action to use. Then check when you last did any of it. …I’ll wait.

The same benchmark data notes that 95% of organizations are now using AI in some form for contract lifecycle management, but only 38% have reached full integration maturity. Which means most organizations have AI that can find the clause, but not a process that ensures anyone acts on it. Buying the CLM was the easy part. Building the post-signature discipline is the actual job.

Source: Ironclad: 2026 Contracting Benchmark Report | Procurement Magazine: AI and CLM Webinar Findings, June 9, 2026

The Token Bill Just Arrived. Corporate America Is Doing a Double Take.

Here is the story that has been building quietly since January and finally broke into the mainstream this month.

TechCrunch reported on June 5th: Uber blew through its entire 2026 AI coding budget by April.

Microsoft revoked its developers’ Claude Code licenses months after enabling them over cost.

A Priceline employee told TechCrunch that a routine Cursor contract renewal came back 4 to 5 times more expensive.

Let’s all be shocked….

The headline number that should make every procurement and finance leader stop scrolling: one company reportedly racked up a $500 million AI bill after failing to enact usage caps for employees. One. Company. Half a billion dollars. Per Elvex’s enterprise token cost analysis, a healthcare enterprise consumed 1 trillion tokens over six months, generating over $6 million in unplanned costs before the finance team even understood what was driving it.

OpenAI CEO Sam Altman said it plainly on stage: “People are really saying, you know, it’s kind of a meme now, but ‘My company spent my entire 2026 budget in Q1. Can you make this more efficient?’ That went from, at the beginning of this year, an issue that never came up, to all of a sudden, a huge issue.”

The structural problem is this: per-token prices have actually fallen. But agentic AI tools burn tokens in chains, not single queries. One autonomous agent completing a sourcing task can fire 40 sequential model calls where a chatbot would fire one. Token consumption scales geometrically with agentic adoption, not linearly. Your 2026 AI budget was probably built on 2025 consumption math. That math is outdated and now wrong.

This is what happens when we plan for something we don’t really understand yet….

If you have agentic AI tools deployed and your finance team is not receiving weekly consumption reports broken down by tool, team, and use case, you are flying blind at altitude. The governance muscle for AI spend is not a future capability. It is a now capability. Token budgets, usage caps per team, and approval workflows for new AI tool rollouts belong in your indirect procurement policy today.

J.R. Storment, executive director of the FinOps Foundation, said companies started calling in April saying they were already three times over their entire 2026 token budget. The whole conversation shifted, in his words, from “tokenmaxxing and ‘go fast’“ to “we need guardrails, how do we control this.” That is the moment indirect procurement walks back into the room. Someone has to own this spend category. It should be you.

Source: TechCrunch: The Token Bill Comes Due, June 5, 2026 | Yahoo Finance / Axios: Companies Evaluate Aggressive AI Spending | Tom’s Hardware: Sam Altman on AI Token Costs, June 2026 | Elvex: AI Token Cost Enterprise Budget Control

Contracts Coming Up? Here Is What to Know About the Vendors You Actually Deal With.

This week hands you specific leverage with four of the biggest indirect spend vendors most of us deal with. Here is what is happening with each of them and how to use it.

Microsoft 365 prices increase July 1, 2026. If you renew before June 30, you can lock in current rates for one year. That is 15 days away. If your renewal is anywhere in the next 12 months, this is worth a conversation with your account team right now.

Separately, GitHub Copilot just moved from per-seat to usage-based billing with AI credit overages. This means your GitHub Copilot costs now have a variable tail that sits on your Azure bill separately from your EA. If you have developers using Copilot Studio or building agents, a single M365 Copilot license at $30 per month can generate $85 or more in additional Azure consumption on top depending on agent traffic. Those costs show up in two different places. You need to look at both.

Tip 1: Microsoft’s fiscal year closes June 30. That is this month. Q4 is the highest-pressure quarter for their sales teams. Buyers who time negotiations to this window consistently see better outcomes. If you are anywhere near a Microsoft renewal, initiate the conversation this week, not next month.

Tip 2: Negotiate Unified Support separately from your EA. Vendr data shows buyers who negotiate support independently reduce those costs by 15 to 30%. It is priced as a percentage of your total Microsoft spend, not based on what you actually use. There is room there.

Tip 3: Before June 30, run a license audit. Copilot Chat enhancements are being added to multiple M365 suites starting this month, which means more users will touch AI features by default. Make sure you are not paying for licenses that are not being used before locking in another year at those quantities.

Source: AppDirect: Microsoft 365 Price Change Advisory, June 2026 | LicenseQ: Microsoft Licensing Update June 2026 | SAMexpert: Microsoft AI Costs Licensing vs Azure

AWS’s Enterprise Discount Program has been rebranded as the Private Pricing Agreement but the commercial structure is identical. AWS’s fiscal year closes December 31, which means you are currently in the middle of the lowest-pressure quarter of their year. That cuts both ways: AWS is less motivated to concede right now, but you are not in a rushed position either. Use Q2 and Q3 to build your leverage case. Do not wait until November.

The AI spend crisis described above creates a specific AWS angle. If you have been over-consuming AI services on AWS and want to renegotiate your commitment, the 60th to 70th percentile of expected consumption is the right sizing posture for your EDP. Optimistic sizing at the 90th percentile creates shortfall risk that AWS will hold against you. Optimize your workloads before you commit, not after.

Tip 1: AWS support is negotiable even though they present it as fixed. Enterprise Support is a revenue line for AWS and is included in EDP negotiations at accounts with meaningful leverage. Always ask. Bundling support into your EDP conversation can save 12 to 18% on the support tier versus negotiating it separately.

Tip 2: Rollover provisions are not in AWS’s default PPA template but are routinely available at signing for buyers who raise them. A rollover converts unused first-year commitment into second-year capacity. If your consumption is uncertain, this provision is worth asking for explicitly.

Tip 3: At $5 million or more in annual AWS spend, mentioning Azure or GCP benchmarks opens doors. Get a real competing quote before your negotiation, not a hypothetical. A documented alternative changes the conversation.

Source: Redress Compliance: AWS EDP Negotiation Guide 2026 | The Negotiation Experts: AWS EDP Discount Tiers and Tactics

Oracle is the vendor most procurement teams dread, and for good reason. They audit aggressively, their licensing rules are intentionally complex, and they have a long institutional memory for customers who tried to walk away. But there are real levers.

The most important thing Oracle does not want you to know is that cross-cloud pricing benchmarks are your most effective negotiation tool. If you are negotiating Oracle Cloud Infrastructure or Oracle SaaS, get a detailed cost estimate from AWS or Azure for equivalent services first. A documented statement that says your analysis shows running equivalent workloads on AWS costs $X and Oracle’s quote is $Y creates a conversation that vague complaints about being overpriced do not.

Tip 1: For Oracle applications like Fusion ERP or HCM, obtain competing proposals from SAP S/4HANA or Workday before you negotiate. You do not have to intend to switch. You need to credibly demonstrate you evaluated alternatives. Oracle responds to documented competitive pressure far more than to verbal negotiating positions.

Tip 2: Oracle’s fiscal year ends May 31. You just missed the highest-leverage window, which means Oracle account teams have already made their numbers and have less urgency to deal right now. Use Q1 of their fiscal year (June to August) to set the terms of a Q4 negotiation, not to close one.

Tip 3: If you are a cloud customer, watch your Processor License Set definitions carefully. Oracle has been aggressive about claiming that virtualized environments require additional licensing. Get written confirmation of your license positions before any infrastructure changes, not after.

Source: Redress Compliance: Oracle Pricing Benchmarks and Negotiation Leverage, May 2026

Salesforce’s fiscal year closes January 31, which means you are entering the period when their Q1 numbers matter and the pressure to sell is moderate but real. The current environment also gives you a specific angle: Salesforce’s Agentforce product is priced at $2 per agent action resolution. That is a consumption model sitting on top of your existing seat-based contracts. Before you add Agentforce to your estate, understand exactly how actions are metered.

The broader Salesforce negotiation dynamic is that the biggest discounts come at the point of most leverage, which is the beginning of the relationship or a major expansion, not renewal. At renewal, your best moves are usage-based: document what you actually consumed versus what you paid for, and build your negotiating position from that data.

Tip 1: Consolidate your Salesforce portfolio into a single negotiation event. If you have Sales Cloud, Service Cloud, Marketing Cloud, Tableau, MuleSoft, and Slack all under different agreements and different account reps, you are losing aggregation leverage. Ask for a single point of contact and a master agreement.

Tip 2: Marketing Cloud renewals are usage negotiations, not seat negotiations. Contact volume and super messages drive the bill. Measure your actual consumption against your contracted tier before you renew. Unused credits in your current contract are your strongest negotiating data point.

Tip 3: For AI products like Agentforce and Einstein, ask for a pilot period with capped consumption before signing a volume commitment. The companies blowing their AI budgets this quarter are the ones who committed to volume before they had real consumption data. Do not be that company.

Source: Everest Group: Nine Tactics for Salesforce Contract Negotiation | Redress Compliance: Salesforce Marketing Cloud Negotiation, June 2026 | Zylo: Salesforce Enterprise License Agreement Guide

Today, June 15 G7 Summit in Evian: Long-term Hormuz governance and nuclear framework talks

June 22 USTR hearing request deadline on Section 301 forced labor tariffs

June 29 Global Procurement Awards AND Global Supply Chain Awards entries close

June 30 Microsoft M365 price lock deadline AND FAR Overhaul Phase 1 changes land

July 6 Written comments close on Section 301 forced labor tariff proposals

July 7 USTR public hearing on forced labor tariffs

July 29 Agency deadline to renegotiate top 10 non-fixed-price federal contracts

That is your week.

The Hormuz deal is the headline but the contract cleanup it requires is the actual work. Agentic AI contracting is the slow-moving problem that will surprise everyone who ignored it. DOGE’s downstream effect on your indirect vendors is underappreciated. Post-signature value leakage is the thing most teams know is real and almost none have a process to address. The token bill crisis is what happens when procurement is not in the room when AI tools get deployed. And the vendor tips section exists because knowing the vendor’s calendar and their pain points is just as important as knowing your own.

Drop a comment.

Tell me where you disagree. Procurement deserves conversations that are honest, not comfortable.

SOURCES

1. Al Jazeera: US-Iran Ceasefire Deal, June 14, 2026

2. Axios: US-Iran Framework Agreement, June 14, 2026

3. CSIS: The Strait of Hormuz in 8 Charts, June 11, 2026

4. PYMNTS: Contracting for Agentic AI Looks Like Outsourcing, February 2026

5. LinkSquares PR Newswire: All-Agentic CLM Launch, May 5, 2026

6. Pramata: CLOC 2026 Recap on Contract Intelligence, June 2026

7. PwC: Agentic AI in Procurement, June 2026

8. Fed-Spend: DOGE Spending Cuts by Agency Tracker, March 2026

9. PBS NewsHour: 40% of DOGE Cancellations Produce No Savings

10. Ironclad: 2026 Contracting Benchmark Report

11. Procurement Magazine: AI and CLM Webinar, June 9, 2026

12. TechCrunch: The Token Bill Comes Due, June 5, 2026

13. Yahoo Finance: Companies Evaluate Aggressive AI Spending

14. Tom’s Hardware: Sam Altman on AI Token Costs, June 2026

15. Elvex: AI Token Cost Enterprise Budget Control

16. AppDirect: Microsoft 365 Price Change Advisory, June 2026

17. LicenseQ: Microsoft Licensing Update June 2026

18. SAMexpert: Microsoft AI Costs: Licensing vs Azure

19. Redress Compliance: AWS EDP Negotiation Guide 2026

20. The Negotiation Experts: AWS EDP Discount Tiers and Tactics

21. Redress Compliance: Oracle Pricing Benchmarks and Negotiation Leverage, May 2026

22. Everest Group: Nine Tactics for Salesforce Contract Negotiation

23. Redress Compliance: Salesforce Marketing Cloud Negotiation, June 2026

24. Zylo: Salesforce Enterprise License Agreement Guide

I’ve been thinking about that sign a lot lately.

Over the last few years, I have spent more time in conversations about how we’re going to use AI to work better and faster than I’ve spent actually working better and faster. I mean, have you been to a conference lately? It almost feels like they’re working on a quota where if they don’t say AI at least 50 times an hour, they have to pay a fine.

The meetings have different titles now. The slide decks are prettier… mainly because people are using Claude to build them. But the output is remarkably unremarkable… a lot of discussion and not a lot of results. It feels like busy work with the sole purpose to just look good.

I don’t think I’m alone in this. I think this is happening everywhere, to everyone, and most people are too busy attending the next AI strategy session to process their thoughts or say it out loud.

Or they’re afraid to be labeled anti-AI, A procurement dinosaur, or just the negative person.

I don’t think I am any of those… I think AI has great potential, probably uncapped potential. I think bad business is its only real hindrance.

So, I’ll say the inside thought outloud… And I’ll bring references.

That’s not a hot take. That’s the conclusion of RAND Corporation research, which also found that AI projects fail at roughly twice the rate of traditional IT projects.¹ S&P Global went further: 42% of companies abandoned most of their AI initiatives in 2025, up from just 17% the year before.² Forty-two percent! Gartner is now predicting that more than 40% of agentic AI projects will be abandoned by 2027 due to unclear ROI, cost overruns, and governance gaps.³

The investment is enormous. The results are, to put it generously, mixed.

IBM’s 2025 CEO study found that only 25% of AI initiatives had delivered expected ROI, and only 16% had scaled enterprise-wide.⁴ Seventy-four percent of companies showed no tangible value from AI investments despite collectively spending $252 billion on it in 2024.⁵ BCG surveyed over a thousand senior executives across 20 sectors and found that only 4% of companies have what they’d describe as cutting-edge AI capabilities. Four percent!

And yet the meetings continue.

Here’s what I think is actually going on.

Companies got excited about a technology before they were remotely ready to use it. One of the downsides to being unmedicated for ADHD is you get very attracted to the shiny object. Rather than admitting that the company lacks medication, they scheduled more meetings. The problem is not that AI doesn’t work. The problem is that most organizations aren’t in any condition to make it work. Their data is a mess, their governance doesn’t exist, their cost controls are nonexistent, and nobody has a clear answer for what “managing this long-term” actually looks like in practice.

Only 7% of enterprises say their data is completely ready for AI, according to a joint study by Harvard Business Review Analytic Services and Cloudera. Seven percent!⁶ Seventy-three percent of organizations say they actively struggle with AI data preparation. Informatica’s 2025 CDO Insights survey found that 43% of organizations cite data quality and readiness as their single biggest obstacle to AI success.⁷ Publicis Sapient put it plainly in their 2026 industry report:

That’s not a technology problem. That’s a homework problem.

The technology showed up and companies realized they hadn’t done any of the foundational work that would make it useful. Rather than doing that work, many of them hired consultants and scheduled workshops and formed AI task forces. Which is exactly what you do when you don’t know what else to do but can’t admit it. Leadership need to look good in front of their peers. Confidence is more important that one may think….

And the governance situation is, if anything, worse. Seventy-eight percent of enterprises are unprepared for their EU AI Act obligations.⁹ Forty-two percent of companies still lack formal governance frameworks for AI data accuracy and control.¹⁰ Gartner estimates that less than 10% of large enterprises had formal AI risk management programs in place as recently as 2023.¹¹ We are deploying systems we don’t fully understand into organizations that haven’t built the controls to manage them. And then we’re writing that off as a strategy.

I keep thinking about RPA.

Remember when robotic process automation was going to change everything?

Every conference had a keynote about it. Every vendor had a platform for it. Every organization had a pilot program. RPA was going to eliminate the repetitive, manual work that was slowing everyone down and free humans up for higher-value tasks. In fairness, RPA did deliver real value in specific, well-defined use cases. Automating invoice processing, handling structured data entry, managing high-volume repetitive tasks where the inputs don’t change.¹²

But most organizations discovered, usually after significant investment, that RPA is fragile. Change the format of an input and the bot breaks. Try to stretch it into complex decision-making, and it falls apart. As of a few years ago, only 13% of large firms had actually industrialized RPA at scale, despite years of enthusiasm and investment. Most were still stuck in pilots.¹³

Does that trajectory sound familiar?

I’m not saying AI is RPA. It’s not. The capability ceiling is genuinely different and the potential is real. But the organizational pattern looks identical. Excitement before readiness. Investment before infrastructure. Strategy sessions before the actual strategy. Then, when results don’t materialize on the timeline the slide deck promised, a slow, quiet pivot to the next thing, with the previous initiative left running somewhere in a corner of the organization that nobody officially manages anymore.

What’s different about AI is the scale of the investment and the speed of the hype cycle. Companies spent $684 billion on AI initiatives globally in 2025.¹⁴ That’s just what people are willing to admit to or had to report. More than $547 billion of that, over 80%, failed to deliver intended business value. That is not a rounding error. That is a structural problem dressed up as innovation strategy.

I want to be clear that my frustration isn’t with the technology itself. My frustration is with the theater surrounding it.

The AI steering committees.

The use case inventories that never get prioritized.

The proof-of-concept that lives in a PowerPoint for six months because nobody will make a decision about whether actually to deploy it.

The vendor demo that becomes a three-month evaluation process that produces a thirty-page report recommending a second evaluation process.

MIT research found that 95% of enterprise generative AI pilots failed to deliver measurable P&L impact, mostly due to integration, data, and governance gaps, not anything wrong with the models themselves.¹⁵ … I mean, companies still struggle implementing things like SAP and Coupa and those are time-tested at this point.

The technology is not the bottleneck. It us… the organization and end users that are the bottleneck. I’ve seen a lot of organizations be very good at scheduling meetings to discuss the bottleneck without actually moving it.

The companies that are getting real results from AI are doing something different.

They’re committing to fewer use cases, not more. They’re investing in data readiness before deployment. They’re building governance structures before they need them, not after something goes wrong. BCG found that leading companies expect 2.1 times greater ROI by focusing on fewer, better-defined use cases rather than spreading AI across every possible function at once.¹⁶

None of that requires a steering committee. It requires someone with authority to say: we are not ready, here is what ready looks like, and here is what we are going to do about it before we have another conversation about use cases. To be the C-level executive that says “we aren’t ready”... That takes a special kind of leader and those ones are rare.

I don’t know if this is a bubble. The investment is real, the underlying technology is genuinely capable, and unlike some past technology cycles, the large language models actually do something impressive. But impressive technology deployed into unprepared organizations with messy data and no governance is not transformation.

It’s an expensive hobby.

The sign is still relevant. We’re going to keep having AI strategy sessions until we figure out why no actual AI work is getting done.

Sources

1. RAND Corporation (2024). AI project failure analysis: more than 80% of AI projects fail, at roughly twice the rate of non-AI IT projects. Referenced in WorkOS, “Why Most Enterprise AI Projects Fail.” https://workos.com/blog/why-most-enterprise-ai-projects-fail-patterns-that-work

2. S&P Global Market Intelligence (2025). Survey of 1,000+ enterprises across North America and Europe: 42% of companies abandoned most AI initiatives in 2025, up from 17% in 2024. Referenced across multiple sources including Neuwark and WorkOS analyses.

3. Gartner (January 2025). Prediction: more than 40% of agentic AI projects will be abandoned by 2027 due to unclear ROI, cost overruns, and governance gaps. Eluminous Technologies, “RPA vs AI in 2026: Enterprise Insights.” https://eluminoustechnologies.com/blog/rpa-vs-ai/

4. IBM CEO Study (May 2025). Only 25% of AI initiatives delivered expected ROI; only 16% scaled enterprise-wide. Neuwark, “Enterprise AI Failure Rate.” https://neuwark.com/blog/enterprise-ai-failure-rate-why-85-percent-of-ai-projects-fail

5. BCG, “Where’s the Value in AI?” (October 2024). Survey of 1,000 CxOs across 59 countries: 74% of companies show no tangible value despite $252.3B in 2024 AI spending; only 4% have cutting-edge AI capabilities. Talyx analysis. https://www.talyx.ai/insights/enterprise-ai-implementation-failure

6. Harvard Business Review Analytic Services and Cloudera (March 2026). “Taming the Complexity of AI Data Readiness.” Only 7% of enterprises say their data is completely ready for AI; 73% struggle with AI data preparation. https://www.cloudera.com/about/news-and-blogs/press-releases/2026-03-05-only-7-percent-of-enterprises-say-their-data-is-completely-ready-for-ai-according-to-new-report-from-cloudera-and-harvard-business-review-analytic-services-reveals.html

7. Informatica CDO Insights Survey (2025). 43% of organizations cite data quality and readiness as top obstacle to AI success. Quick Launch Analytics. https://quicklaunchanalytics.com/bi-blog/why-80-of-ai-projects-fail-before-they-start-its-your-data-foundation/

8. Publicis Sapient, 2026 Guide to Next Industry Trends Report (November 2025). “AI won’t fail for lack of models. It will fail for lack of data discipline.” https://ppc.land/data-governance-gap-exposes-ai-confidence-crisis-across-industries/

9. Vision Compliance / Optro AI Governance Stats (2026). 78% of enterprises unprepared for EU AI Act obligations. https://optro.ai/blog/ai-governance-stats

10. Clari Labs Research (January 2026). 42% of enterprises still lack formal governance frameworks; 87% of enterprises missed 2025 revenue targets despite record AI investment. https://www.businesswire.com/news/home/20260114516767/en/New-Clari-Labs-Research-Reveals-87-of-Enterprises-Missed-Revenue-Targets-in-2025-Despite-Record-AI-Investment

11. Gartner prediction: by 2026, 50% of large enterprises will have formal AI risk management programs, up from less than 10% in 2023. Quinnox, “Data Governance for AI in 2025.” https://www.quinnox.com/blogs/data-governance-for-ai/

12. RPA use cases and limitations: Linford Co, “Is RPA Dead?” (February 2025). https://linfordco.com/blog/is-rpa-dead/

13. Only 13% of large firms had industrialized RPA at scale. Neuronimbus, “The Future of RPA in 2025-2027.” https://www.neuronimbus.com/blog/the-future-of-rpa-robotic-process-automation-in-2025-2027/

14. Global enterprise AI investment of $684 billion in 2025; 80%+ failed to deliver intended business value. Pertama Partners, “AI Project Failure Statistics 2026.” https://www.pertamapartners.com/insights/ai-project-failure-statistics-2026

15. MIT Project NANDA (July 2025). 95% of enterprise generative AI pilots delivered zero measurable P&L impact, primarily due to integration, data, and governance gaps. SR Analytics. https://sranalytics.io/blog/why-95-of-ai-projects-fail/

16. BCG AI Radar. Leading companies focusing on fewer use cases expect 2.1x greater ROI. Neuwark, “Enterprise AI Failure Rate.” https://neuwark.com/blog/enterprise-ai-failure-rate-why-85-percent-of-ai-projects-fail

There’s a big one though. More of us suffer from it than we realize…. It’s a real issue nobody discusses but I’m going to right now.

There’s a skill gap in legal and procurement that nobody wants to talk about out loud.

Why? Personally, I think it’s because companies don’t know what it’s costing them in labor cost to deal with the knowledge gap.

It’s not negotiation tactics. It’s not contract structure. It’s not even knowing the difference between indemnification and limitation of liability…

It’s this: too many attorneys, new and experienced alike, are reviewing contracts for things they don’t actually understand and you CAN’T redline what you don’t understand. More specifically, you cannot assess risk in something you don’t understand and, let’s be real, contract review, at its core, is risk assessment.

That’s it. That’s the whole job. ¹

If you don’t know what you’re assessing, you’re not doing risk assessment. You’re doing an off-Broadway theater production.

Here’s what contract review looks like in a lot of organizations…. legal gets handed an agreement, opens a clean copy, and starts marking it up. Liability caps. IP ownership. Termination provisions. Auto-renewal clauses. The usual suspects get flagged, the unusual ones get missed, and the whole thing goes back to the vendor looking busy.

Everybody feels good because it looks like work got done. A day well spent because a contract went from black to red.

The problem is, nobody asked the most important question first: What are we actually buying, and what can actually go wrong with it?

I’ve seen 30-year attorneys redline staffing agreements the same way they redline software licenses. I’ve seen junior associates spend three hours tightening up an indemnification clause in a contract for a service they couldn’t define if you put a billable hour on the line. The red ink flows. The deal moves forward and somewhere downstream, something breaks that a better question at the beginning would have caught.

The boilerplate indemnification that works fine when you’re buying marketing services doesn’t work the same way when you’re buying a SaaS platform sitting inside your network touching employee data. Neither of those works the same way when you’re buying a jet.

This isn’t a dig at attorneys….and as an attorney I am allowed to take shots at the group. It’s a dig at how I was trained … how we’re training them, or more accurately, not training them.

Let me give you a real example of what I mean.

At one point in my career, I had to buy a corporate jet. I want to be clear about my qualifications going into this: I know how to purchase a ticket, find my seat, and pick a movie. That’s the full extent of my aviation expertise. I’m also real good at booking the Uber and securing in-flight snacks on my corporate card.

So, I had to learn. Fast…

I had to understand who was actually flying, which executives, which clients, which routes, and map out the typical destinations. Why? Because the plane that gets you from a small regional airport outside Boston to a small regional airport outside Baltimore is not the same plane that gets you from Boston to LAX. And it’s definitely not the plane that gets you to Heathrow. Runway length. Range. Fuel capacity. Cabin configuration for long-haul versus short-haul. These are not minor details. They are the entire point.

Ironically enough, the most important detail was the inflight service and at what temperature the water would be served. One thing I learned is that services are a separate RFP.

I had to understand what I didn’t know, and then go learn it, because you cannot add things to a contract when you don’t know they exist. You can’t protect against a risk you haven’t identified. Brushing up the indemnification clause and adding standard insurance requirements doesn’t cover you when the specific aircraft can’t land at the specific airports your company actually uses. That’s not a legal problem. That’s a you-didn’t-do-your-homework problem.

The contract for a corporate aircraft has to address maintenance obligations, inspection schedules, crew qualifications, operational limitations, storage, and insurance that actually reflects aviation risk… deicing, storage, and so much more. None of which looks anything like what you’d put in a services agreement. If you open that contract thinking you can rely on your standard playbook, you are not protecting your client or your company. You’re coloring on a page and calling it contract review.

Now take staffing. It seems simple until you realize how many different things the phrase “HR vendor” can mean, and how dramatically different the risk exposure is depending on which one you’re actually buying.

Temp staffing, contingent search, and retained search all fall under the same general umbrella. They all involve bringing humans into your organization. They all have similar-looking commercial terms on the surface. Yet, they carry fundamentally different risk profiles, and if you’re reviewing them all the same way, you’re not doing your job.

This has been a pet peeve of mine for years. I’ve worked with so many attorneys who do this work but don’t understand it. They operate with such confidence. …typical.

Let’s talk about it … it’s a real good example…

With temp staffing, the worker is on the vendor’s payroll. You’re buying labor hours. The risk assessment here is not primarily about rates. It’s about co-employment liability. It’s about worker classification. It’s about what happens if that person is injured on your site and who is actually responsible for taxes, benefits, and wage compliance. ² It’s about replacing your temp employee as you see fit. Your indemnification language needs to clearly establish that this person is not your employee, and it needs to hold up when something goes wrong, not just read well when nothing has. Microsoft learned this the hard way when a class-action lawsuit over temp worker benefits took eight years to resolve and cost the company $97 million. ³ That is not a hypothetical risk. That is a real number attached to real language that somebody thought was adequate until it wasn’t.

Contingent search is a transaction. You pay a fee when a candidate gets hired. The vendor’s obligation ends at placement, kind of, but if you know what you’re doing you contract appropriately. The risk here lives in the guarantee period, specifically what “replacement” means, and what happens at day 89 when your guarantee runs 90. Because that candidate will be gone on day 91 and you’ll be back at square one arguing about contract language nobody thought hard enough about upfront.

Retained search is a relationship with a fee structure attached to process, not outcome. You’re paying the fee regardless which is really hard for some attorneys to rationalize. You’re paying regardless of whether someone gets hired. The risk calculus is entirely different: what do you actually get if the search fails? What does exclusivity mean and how is it enforced? What’s the recourse when the firm delivers three finalists you already knew about? What happens if you’re selected candidate joines and backs out in a few weeks?

Same general category. Same section of your vendor master. Completely different contracts. Completely different risks. If you pick up the pen without knowing which one you’re reviewing, you’re pattern-matching against language you’ve seen before and hoping for the best.

Now consider IT, which is where the stakes get genuinely frightening and botching it gets real expensive real quick.

Software agreements have become some of the most consequential contracts a company signs, and a lot of attorneys still review them like they’re boilerplate. They’re not. The risk in a bad software contract can be staggering, and almost none of it lives in the clauses you were trained to flag.

Before you touch a software or SaaS agreement, you need to understand what you’re actually buying AND how it will work in action. Not “it’s a project management tool.” You need to ask lots of questions… What data does it touch? What workflows does it sit in? How does it connect to your existing systems? And where does the data actually go?

That last question matters more than most attorneys realize. “The cloud” is not an answer. Whose cloud? What region? What do their subprocessors look like and do they change without notice? There is a meaningful difference between data sitting on a server in Virginia and data sitting on a server in Frankfurt, especially if that data is subject to HIPAA, GDPR, or your own contractual obligations to customers. GDPR violations alone can cost up to 4% of global annual revenue. ⁴ HIPAA penalties run up to $1.5 million per year per violation category. ^5. These aren’t edge cases. They are the predictable consequence of signing a contract without understanding what’s in the system.

And what kind of data is going in? There is a massive difference between a tool that stores project timelines and a tool that stores employee health information, customer financial records, or proprietary business processes. The average cost of a data breach in the technology and SaaS sector is now $5.3 million per incident. ⁶ Healthcare breaches average $9.77 million. ⁷ The limitation of liability cap that seemed reasonable for a productivity tool is laughably inadequate when you’re talking about a breach that exposes regulated data. Now you’re balancing liability cap and insurance.

If you don’t understand how the system works and what’s in it, your data security provisions are guesswork. Your SLA is measuring uptime on a system you don’t actually understand. Your audit rights might be perfectly drafted to cover the wrong things entirely. Now, you just wait until your vendor is hacked and your info, along with all their clients, is exposed.

Here’s the uncomfortable part. Law schools teach legal reasoning. That’s not wrong, it’s just incomplete. Somewhere along the way, firms and legal departments decided the answer was to bring in specialists or outside counsel when things got technical. Which means the attorneys doing first-pass review have never been told they need to understand the thing before they can protect against the risks in the thing.

The senior attorneys aren’t always helping. Some of the heaviest redliners I’ve seen are experienced lawyers who learned contract review when deals were simpler, or when the technical complexity could be safely ignored. They found the clauses they knew to look for, marked them up, and called it done. Then they trained the next generation to do exactly the same thing.

The result is a profession where contract review often means pattern-matching against known clauses rather than actually identifying risk in the specific transaction in front of you. That works fine, right up until someone hands you a jet purchase agreement and you reach for your standard services playbook.

Before any agreement gets reviewed, there needs to be a real conversation. Not a long one. But an honest one.

We need to teach attorneys to ask questions to understand what they don’t know….

What are we buying? Who asked for it? What does the business actually do with this thing? What happens if the vendor disappears tomorrow? What are the actual failure modes, not the theoretical ones, the real ones specific to this deal?

The attorney’s job is to translate those answers into contract language that protects against the real risks, not the theoretical risks, not the risks from the last deal that kind of looked like this one, but the actual risks in this specific transaction.

Gone are the days when all redlines are created equal. To be good at this job, you need to know what to add just as much as you know what to change in so many different circumstances.

Redlining is the easy part. Knowing what to protect is hard. And right now, we’re teaching the easy part and calling it contract review.

Sources

1. “The highest level of work is to assess risk associated with contract clauses and advise on solutions. This risk assessment work is highly contextual and depends on the industry, the business model, the risk tolerance and the priorities of a company.” CUAD: An Expert-Annotated NLP Dataset for Legal Contract Review, arXiv (2021). https://arxiv.org/pdf/2103.06268

2. Co-Employment Risk, CXC Global (2025). https://www.cxcglobal.com/glossary/co-employment-risk/

3. Vizcaino v. Microsoft Corp. co-employment class action; $97 million settlement. See: Randstad Enterprise / Staffing Industry Analysts coverage of SIA co-employment liability laws. https://www.randstadenterprise.com/insights/randstad-enterprise-insights/sia-understanding-co-employment-liability-laws-2/

4. GDPR penalties: up to €20 million or 4% of global annual turnover, whichever is higher. European authorities issued over €3 billion in GDPR fines since enforcement began. Secure Privacy, SaaS Privacy Compliance Requirements 2025 Guide. https://secureprivacy.ai/blog/saas-privacy-compliance-requirements-2025-guide

5. HIPAA penalty structure: up to $50,000 per violation and $1.5 million per year for willful neglect. Appinventiv, Cloud Compliance Requirements (2026). https://appinventiv.com/blog/cloud-regulatory-compliances-guide/

6. Technology & SaaS sector average breach cost: USD $5.3 million; cloud misconfiguration implicated in 29% of breaches. DataStackHub, Data Breach Statistics 2025–2026. https://www.datastackhub.com/insights/data-breach-statistics/

7. Healthcare data breach average cost: $9.77 million. Callidus AI, The Data Privacy & Security Clauses in SaaS Agreements Attorneys Can’t Overlook (2025). https://www.callidusai.com/data-privacy-security-clauses-in-saas-agreements/

The FAR Is Being Rewritten And You Are Running Out of Time to Care

Let us get straight to it. The Federal Acquisition Regulation, which is the rulebook that governs literally everything the federal government buys, is being gutted and rebuilt right now. Not someday… right now. The official name for this chaos is the Revolutionary FAR Overhaul, which is the kind of name only this government leadership team could come up with and say with a straight face.

On April 30th, President Trump signed an Executive Order titled Promoting Efficiency, Accountability, and Performance in Federal Contracting. The short version is this: fixed-price contracts are now the default. Cost-reimbursement contracts, time and materials, labor hour, all of it now requires written justification from the contracting officer to the agency head just to use. This is a significant shift. The administration’s stated reasoning is that cost-plus contracting has produced “unpredictable costs, bloated overhead, and weak performance incentives.” Whether you agree or not, the policy is real, it is moving fast, and it has immediate consequences.

If you currently hold a large federal contract on a cost-reimbursement basis, you should expect to hear from your contracting officer soon. If you are a subcontractor to a prime on one of those contracts, your prime is almost certainly about to try to pass that pain downstream to you. Understand your rights before anyone comes knocking.

Here are the live deadlines that should be on your calendar right now:

June 14, 2026: OMB must issue implementation guidance to all agencies on the fixed-price EO

July 29, 2026: Every federal agency must review its 10 largest non-fixed-price contracts and seek to renegotiate them to incorporate fixed prices

August 28, 2026: Proposed FAR amendments must be submitted to permanently codify fixed-price contracting as the default

June 30, 2026: Phase 1 of the FAR Overhaul’s biggest interim rule changes land

And no, the private sector is not immune. If you sell to the government, source from government contractors, or have any slice of revenue touching federal contracts, this affects your world.

Source: Arnold & Porter Client Advisory, May 2026 | PilieroMazza Law, May 2026 | Federal News Network, May 2026

USTR Just Declared 60 Countries Guilty of Enabling Forced Labor. Your Sourcing Strategy Has 28 Days.

On June 2nd, the Office of the United States Trade Representative concluded its Section 301 investigations into 60 economies simultaneously and found that every single one of them had failed to properly ban imports made with forced labor. Every one. Zero countries passed.

The proposed punishment: an additional 10% tariff on imports from 15 trading partners including Bangladesh, the UK, Canada, Mexico, and the EU, and 12.5% on the remaining 45, which includes China, India, Vietnam, South Korea, and Australia. These tariffs are not in effect yet. But the comment window closes July 6th and the public hearing is July 7th, which means the clock to push back or protect yourself is genuinely short.

Map your exposure by country of origin. Pull your 2025 and 2026 entry data for all 60 economies. Model both duty layers on your margin. Review your tariff pass-through clauses in supplier contracts. If your sourcing touches any of these countries and your contracts do not have force majeure or tariff adjustment language, that is a gap you want to close before this lands.

One more wrinkle worth watching: a temporary 10% import surcharge under Section 122 expires on July 24th unless Congress extends it. USTR is widely expected to try to get the Section 301 forced labor duties in place before that date to avoid any gap in tariff coverage. The pace here is not slow.

And separately, on June 1st, Trump issued a proclamation modifying Section 232 duties on aluminum, steel, and copper. And Canada formally pushed for USMCA renewal for another 16 years ahead of the July review. Trade policy is moving faster right now than most sourcing teams can track. If that is you, that is not an insult, that is the situation, and it is a reasonable response to an unreasonable pace of change.

Source: USTR Official Press Release, June 2, 2026 | Kesco Logistics Weekly Freight Report, June 5, 2026 | Clark Hill Trade Alert, June 2026

93% of Organizations Cannot Predict a Supply Chain Disruption & the Other 7% are Lying.

ProcureAbility, a Jabil company, dropped a new report this week titled From Insight to Impact: Analytics-Led Third-Party Risk Performance Management and the headline finding is not good. Only 7% of organizations have mature third-party risk performance management systems that can actually model future scenarios. Seven percent. The other 93% are essentially flying blind and hoping nothing catches fire.

This is not a new problem but it is a newly quantified one and that is what people wearing suits are interested in. Traditional vendor questionnaires and annual assessments show you what a supplier looked like at one point in time. They do not show you what that supplier looks like today, whether they patched a critical vulnerability last week, whether their subcontractor just got hit, or whether their cloud provider is the same one three of your other vendors depend on. The threat landscape changes faster than the review cycle.

The instinct is to add more questionnaires. The actual move is continuous monitoring. If your TPRM program is still running on annual assessments and a spreadsheet, you are not managing risk. You are documenting your exposure after the fact.

There is also a broader trend worth tracking here. Supply chain cyber attacks in 2026 have been particularly ugly. Earlier this spring, a trojanized software installer hit organizations across more than 100 countries, a certificate authority was breached through a customer support chat, and a popular open-source framework was compromised in a way that reached inside one of the world’s largest AI companies. These are not exotic attacks. They are trust exploits, attacks that abuse the relationships your organization has with the tools and vendors you already trust.

The question worth asking your team this week is not whether you have a TPRM program. It is whether your TPRM program would have caught any of those three incidents before you read about them in the news.

Source: ProcureAbility PR Newswire, June 3, 2026 | WardenBit Blog, June 2026

Shadow AI Is the New Shadow IT and It Is Already on Your Company Card

Remember the shadow IT era? When people were spinning up Dropbox accounts, buying their own software licenses on corporate cards, and procurement found out months later during an invoice audit? That same story is playing out again, only faster, more expensive, and with a much worse compliance tail.

Shadow AI usage inside enterprises is up 156% since 2023. Unsanctioned AI tools are sitting on company cards for an average of 400 days before anyone catches them. Let that number sit with you for a second. Over a year of undetected spend per tool and these are not cheap tools. Gartner pegs worldwide AI spending at $2.5 trillion this year. Zylo’s 2026 SaaS Management Index shows AI-native app spending up 393% inside large enterprises. 63% of enterprises blow past their AI budget by 30% or more in year one. 78% of IT leaders have already been surprised by an unexpected consumption charge. … Almost like they didn’t know what they were buying or how it worked…

If your indirect team does not have a current view of what AI tools are on your company cards right now, you do not have a spend management program. You have a discovery project waiting to happen.

And the pricing model problem is making this harder. Traditional per-seat SaaS pricing is being replaced, slowly and unevenly, with usage-based and outcome-based models. This matters for indirect sourcing because you can no longer just look at headcount to forecast software costs. An AI agent that resolves 80% of your IT service tickets is not a seat. It is a consumption curve you need to model. A useful gut-check for every renewal right now: strip the AI marketing layer out completely. If you removed the word AI from every slide and demo, would the underlying workflow still justify the price? If the answer is no, you have a wrapper. Do not renew at last year’s rate.

One major consolidation move to know about: as of June 2026, Vendr is now part of Vertice, combining benchmark data from over 130,000 software transactions with $75 billion in indirect spend across 32,000 vendors. The pitch is AI-powered contract negotiation at enterprise scale with actual market pricing data behind it. Worth watching as a signal of where the indirect intelligence market is heading.

Source: Torii HQ AI Spend Management Report, June 2026 | Zylo 2026 SaaS Management Index | Suplari Blog, June 3, 2026

The S2P Leaderboard Just Updated. Here Is Who Is Winning.

The 2026 Gartner Magic Quadrant for Source-to-Pay Suites is out and the standings are worth knowing. Coupa took the top ranking for ability to execute among all 13 providers evaluated. Zip has grown significantly since its founding and is increasingly showing up as the intake layer sitting on top of SAP Ariba or Coupa rather than replacing them. And SAP is not standing still.

SAP just made next-gen Ariba generally available. This is not a rebrand. It is a rebuilt platform 20 years in the maxking with Joule, SAP’s AI copilot, embedded directly into procurement workflows for bid analysis, supplier response summaries, intelligent contracting, and invoice creation. For organizations already running SAP Cloud ERP, this is a meaningful development because the data flows natively across the source-to-pay lifecycle without middleware patches.

The Zip CEO has a bold public line worth knowing: 90% of procurement tasks automated by 2030, with the CPO evolving into a proactive strategic advisor. That is a vision statement, not a product feature. But it tells you where the intake and orchestration space is heading and what the pitch to your CFO is going to sound like in your next platform review.

The platform wars are no longer about features. They are about which AI architecture you are betting your procurement operations on for the next decade. That is a procurement decision masquerading as an IT procurement decision. Own it.

Source: Procurement Magazine, Gartner MQ Analysis, February 2026 | SAP News Center, March 2026

Next 4 Weeks

June 14 OMB implementation guidance due to agencies on fixed-price contracting EO

June 15 to 16 | Austin TX ProcureCon Connect, the invite-only CPO summit

June 22 USTR hearing request deadline on forced labor Section 301 tariffs

June 29 Global Procurement Awards entries close

June 30 FAR Overhaul Phase 1 major rule changes land

July 6 Written comments close on Section 301 forced labor tariff proposals

July 7 USTR public hearing on forced labor tariffs

July 29 Agency deadline to renegotiate top 10 non-fixed-price contracts

That is your week.

Five stories, all live, all with real deadlines and real dollar consequences. The FAR overhaul and the forced labor tariff wave are the two that need your attention first because both have hard deadlines in the next 30 days and both have the potential to materially change the terms of contracts you already have in place.

If any of this sparked a take, a question, or a good argument, drop it in the comments. Procurement is not a spectator sport.

SOURCES

1. Arnold & Porter: Executive Order on Fixed-Price Contracting arnoldporter.com

2. PilieroMazza: Fixed-Price EO Analysis pilieromazza.com

3. Federal News Network: Fixed-Price Preference Boost federalnewsnetwork.com

4. USTR: Section 301 Forced Labor Findings, June 2, 2026 ustr.gov

5. Kesco Logistics: Weekly Freight Report, June 5, 2026 kescologistics.com

6. Clark Hill: USTR Section 301 Forced Labor Tariff Alert clarkhill.com

7. ProcureAbility / PR Newswire: TPRM Report, June 3, 2026 prnewswire.com

8. WardenBit: Supply Chain Is the New Front Door, June 2026 wardenbit.com

9. Torii HQ: AI Spend Management Tools, June 2026 toriihq.com

10. Zylo: 2026 SaaS Management Index zylo.com

11. Suplari: SaaS Procurement in the AI Era, June 3, 2026 suplari.com

12. Procurement Magazine: Gartner MQ S2P Analysis, Feb 2026 procurementmag.com

13. SAP News Center: Next-Gen SAP Ariba Launch, March 2026 news.sap.com

As procurement and legal folks, we are great storytellers. We have a flair for words… we spin stories every day for our stakeholders, our managers, our direct reports, and our C-level suite of overlords. It’s how we make magic and get people to do the right thing according to us. This is a double-edged sword, too. We sometimes fall victim to our own folklore.

The big story that lives in our world like one of Aesop’s fables. The dinosaur that hired your boss told them the story because it was told to them as a newbie. Then, your boss told you the same story. It goes like this: the big bad vendor (assume AWS / SAP type vendors) has the power and you shouldn’t waste your time. Their high-powered three-piece suit lawyers wrote the contract. Their team has negotiated a thousand of these and bigger players than us have been unable to get them to move. We’re one small deal on their pipeline, and if you push too hard, they’ll just walk to the next buyer or they’ll just draw a hard line for take it or leave it.

That figure comes from practitioners who spent years on the vendor side watching buyers roll over.^[1] Read that again. Eighty-two percent. The list price you received was not the best price… it was an opening bid dressed up in a PDF. Vendors bake 20 to 40% margin into every quote because they’ve learned that most of us don’t push.[1] When you don’t push, they mock you and celebrate on your dime after you sign the deal. That money just stays on their side of the ledger, funding someone’s club level seats at the game.

And before anyone tunes out because they don’t buy SaaS… this isn’t just a SaaS problem. It’s a distributor problem, a staffing firm problem, a logistics problem, an any-category-with-three-vendors-who-matter problem. The dynamic is always the same: a handful of large vendors, a crowd of mid-market buyers, and a quiet group agreement that the buyers have no power.

Spoiler: we do.

Three myths get repeated in every procurement Slack channel I’ve ever been in. Let’s drag them into the light.

Myth 1: “They don’t negotiate with companies our size.”

They do. The rep handling your account has a quota, and quotas are truly, madly, deeply, embarrassingly important to their future with the company. At the end of the quarter and the end of the fiscal year, the pressure to close is real regardless of deal size. A $40,000 contract that closes in the last week of Q4 is better for a rep than a $40,000 contract that doesn’t, and it’s a lot better than a $40,000 contract that closes the following Tuesday. Use that.

Myth 2: “If we push on terms, we’ll damage the relationship.”

Professional negotiation doesn’t damage relationships. It sets the tone for one. Vendors respect buyers who know what they want and can say it out loud. What actually damages relationships is signing a bad deal, resenting it for 24 months, and then sending passive-aggressive emails to your AE about it. You agreed to those terms. The vendor remembers, the contract remembers, and the North doesn’t forget.

Myth 3: “Their standard contract is the contract.”

The paper a vendor sends you is not a ceiling. It’s a starting point they prefer, and the word “standard” stamped on the MSA is doing a lot of heavy lifting.[2] The clause limiting your ability to exit, the auto-renewal notice buried on page 14, the unilateral pricing adjustment language that lets them raise your bill while you sleep… none of that is immovable. It just requires someone in your org to open the document and start marking it up. Everything is up for negotiation and everything has a price but it’s important to remember that price isn’t always money.

Redlining requires leverage, and you probably have more of it than you think, in more places than you’re looking.

Timing is your most underused lever. Vendors close their books quarterly and annually. Procurement teams rarely exploit this. If you’re starting a sourcing process in January for a March go-live, you’re walking into a negotiation where the vendor has no urgency. Start the process so you can close in the last two weeks of a quarter.^[3] The rep’s urgency becomes your negotiating room.

Competitive tension is real, even when you don’t intend to switch. You don’t have to actually plan to buy from Vendor B to use Vendor B in a negotiation with Vendor A. You have to be credible about the possibility.^[4] That means doing enough evaluation that you can discuss what Vendor B offers. Vendors can tell when you’ve done zero diligence on alternatives. They can also tell when you have. The second version of that conversation goes differently.

Usage data is leverage, especially at renewal. If you’re underutilizing licenses, seats, or capacity, that’s not something to hide. It’s something to use.^[5] Walking into a renewal with a usage report showing 60% adoption and asking for a reduction is a data-driven conversation, not a confrontation. Vendors would rather right-size a contract than lose it. You need to know what you’re talking about when you go into a room.

Multi-year commitments move the needle when you’re willing to make them. Vendors want revenue visibility. A two or three-year commitment is worth something to them. Use it to extract price caps, implementation credits, or better SLA terms rather than just a discount.^[1] Trading a multi-year commitment for a discount alone is leaving value on the table.

Which brings us to the contract itself… the part most procurement teams skim because it’s long and the lawyers will “handle it,” although in my experience most contracts in the big corporate world never reach an attorney since they assume procurement has it under control. We spend our energy haggling over price and then sign away terms that quietly decide whether the deal actually works.

Termination for convenience matters more than you think. The vendor’s default paper often ties you to a termination-for-cause standard, which is a polite way of saying you can’t leave unless you can prove they broke the deal. Push for termination for convenience with reasonable notice.[6] Without it, you’re locked in even if performance quietly degrades, the product pivots into something you didn’t buy, or your business simply moves on.

Auto-renewal notice windows are a quiet trap with a very loud bill at the end of it. The vendor-favorable version gives you 30 days to cancel, which is barely enough time to find the contract, let alone read it. The buyer-favorable version gives you 90 to 180 days.[6] Push the window out. It costs the vendor very little and saves you from waking up to a renewal you never meant to sign. Or, if you’re at one of those companies that doesn’t adequately track its notice periods and other dates (you know who you are!), get rid of it completely. If you want a long-term deal, make one. Otherwise, end it in a year and force everyone back to the table. If it’s not a long-term deal… no reason to risk an unbudgeted adventure.

Price adjustment caps protect you in multi-year deals, without one, you haven’t locked in value, you’ve locked in someone else’s pricing strategy. Cap annual increases at 3 to 5%.[1] Vendors will push back, usually with a line about “market conditions.” Hold the line, or trade it for something else worth having. I’ve also had fairly good luck with 3-year deals staying flat. You don’t know if you don’t ask.

Data portability and exit provisions are non-negotiable in SaaS. Before you sign, you should know exactly what happens to your data on the way out the door: format, timeline, cost of extraction.[6] Vendors who make exit difficult aren’t being careless. They’re pricing in switching costs they fully intend to collect later. It’s the reason SAP Ariba has prisoners instead of customers.

Underneath all of these tactics is a deeper issue, and it’s the one nobody wants to say out loud. We often accept bad terms because we’ve already decided the vendor is a “partner,” and negotiating with a partner feels rude. It isn’t.

The best vendor relationships are built on clear expectations, not deference.^[7] A vendor who wins a negotiated deal knows what you expect. They know your exit rights, your renewal process, your pricing expectations. That clarity is good for both sides. The worst vendor relationships are built on ambiguity, usually because the buyer signed paper they didn’t fully negotiate.

You are not a small buyer begging a large vendor for favorable terms. You are a buyer with a budget, a decision to make, and the ability to walk that decision across the street. Act like it.

We should be training for negotiations as hard as fighters do for the next MMA match. We should know our opponent and what is important to them, when their fiscal year ends, and what the dealbreakers for them are and why.

Procurement has spent the last decade trying to earn a seat at the strategic table, and every conference keynote on the subject says basically the same thing… after they finish shaming the room for not using enough AI. Here’s what the keynotes leave out: part of earning that seat is proving we can actually negotiate, not just run an RFP. The unglamorous tactical work of getting a good contract is where that proof lives.

Stop accepting the myth that size determines leverage. SIZE DOESN’T MATTER. Start negotiating like you have options, because you do.

Sources

[1] SaaS Contract Negotiation: Save 20-40% on Software (Softabase, 2026)

[2] Mastering SaaS Contract Negotiations: Proven Tactics for Success (VendorSage, 2025)

[3] Top 12 Tips For A Successful Supplier Negotiation Strategy In 2026 (Spendflo, 2026)

[4] SaaS Contract Negotiation Guide (Aline, 2026)

[5] 2026 SaaS Procurement Trends (Tropic, 2026)

[6] 7 SaaS Contract Negotiation Strategies to Cut Costs (CloudEagle via Medium, 2025)

[7] 40 Must-Know Contract Management Trends for 2026 (Fynk, 2026)

“90% of jobs can be taught. Give people a chance.”

Scrolling LinkedIn, stuff like this comes up all the time. Every time, it gets me. It frustrates me to know. I always think, yeah but not procurement. We’d never allow that.

Some may not agree, and I’m open to being wrong but I wholeheartedly believe Procurement has a gatekeeping problem.

A serious one.

We have built an entire interview infrastructure designed not to find great people, but to eliminate anyone who hasn’t already done the exact job we’re trying to hire for.

Which, if you think about it for more than thirty seconds, is insane.

Here’s the thing about hard skills: I can teach them.

Excel? I can teach that.

Negotiation tactics? Absolutely.

How to run a sourcing event, build a category strategy, read a contract, navigate an RFP process? All teachable.

All of it. Give me someone halfway motivated and I’ll have them fully functional in a couple months.

You know what I cannot teach?

How to be likeable...

Sure, I may sound like a jerk right now but it’s true.

I cannot teach you how to walk into a meeting and not make everyone in the room immediately regret being there.

I cannot teach you how to read a conversation, pick up on what’s not being said, and adjust accordingly on the fly.

I cannot teach emotional intelligence to someone who has spent so many years being completely unbothered by its absence.

I cannot teach… likability.

I cannot teach self-awareness to someone who has never once considered that they might be the problem.

And yet. When we post a job, what are we screening for?

A college degree. Five years of experience. Category management background preferred. Familiarity with Ariba, Coupa, and whatever platform we bought during COVID and never fully implemented. Must have managed suppliers in our industry, at our spend level, in our specific type of organization.

So let me tell you about a hire I made.

This guy had one employer for his entire adult career. Twenty-something years, same company, zero procurement experience. By every standard rubric we use in this profession, he should never have even gotten a callback. This would have been an easy auto screen out by even the most basic ATS tool.

I brought them in anyway.

Now, it’s worth noting that I did take some heat for this. He didn’t have a college degree. He didn’t look like a procurement candidate on paper. I had to put my rear end on the line to make this work.

Within two years, he had grown into the role in a way that made some of my more “experienced” hires look like they were still figuring out which way the building faced. Not because he was some kind of procurement prodigy. He wasn’t. And I’m certainly not some visionary leader who saw what others couldn’t…. or maybe I am - verdict is still out.

What he had was simpler than that. He listened.

Actually listened... not the kind of listening where you’re just waiting for your turn to talk. He learned without ego. He communicated clearly. He gave a genuine damn about the people on the other side of every conversation, whether that was a stakeholder, a supplier, or someone on the team.

That’s it. That was the whole thing. ...This is what should be in job descriptions but never will be…

I don’t think these are exotic qualities. They’re not rare gifts handed out to a lucky few. They’re just... not what we screen for. Because you can’t put them in a job description without sounding like a fortune cookie, you can’t verify them with a resume, and you can set up some ATS or AI tool to weed out “the disqualified” based on keywords.

So, instead we ask for five years of category experience and pat ourselves on the back for being rigorous. Check the box and move on.

The best procurement people I’ve worked with came in sideways and typically by accident. They slide in from Finance or Legal. Some came from the dark side… the vendor side. This guy,… from one employer for twenty years outside of anything that even resembled procurement and contracting. They all figured it out because they were curious, coachable, and, the most critical element, people actually wanted to work with them.

That last part matters more than we admit.

Procurement is a relationship function dressed up in spreadsheets. No matter how pretty you look in your sheets... People need to like you or nobody is going to bother with you and your sheets. You can have perfect process discipline and be completely useless if no one trusts you, no one wants to call you, and every stakeholder you touch is quietly routing around you.

Moral of the story… hire for the thing you can’t fix and teach the rest.

Have a good story about a hire that doesn’t fit the typical mold? Share it with the world. Let’s make Procurement more accessible to quality candidates!

It was never a phase, mom! (Bonus points if you get the reference.)

Then, like many, I got to an age where I had no idea what I wanted to be when I grew up. Hell, I didn’t want to grow up (still don’t!).

I wanted to be all kinds of things over the years. Looking back, I can’t recall a single time I thought to myself... “I want to buy stuff for companies and help them just be better all around”.

Nothing came close to even resembling that.

I didn’t even know procurement was a career. If someone had asked 22 year old me about Procurement, I wouldn’t have even known what it was or where it happened.

There was no Procurement table at the Career fair in High School. It wasn’t something I had ever heard anyone talk about. I wanted to be a police officer. I went to school and earned a degree in criminal justice, fully expecting that to be my path. While in School, I worked for the Commonwealth at the Department of Youth Services in a secure facility for juveniles awaiting trial. It was wild. I learned a lot about how to survive that I still use today at my desk at home.

Anyways, attorneys would come in to represent these children and a lot of them sucked. They treated these kids like absolute garbage which was even worse because the world had already done a fairly good job at that. I decided that law school was my path and I would help kids that nobody else wanted to help.

I became a lawyer and started working in a law firm. Somewhere along the way I took a hard turn from child advocate to focusing on contracts, disputes, and the kinds of legal issues businesses deal with every day.

Procurement was invisible to me.

Until, eventually, I found procurement the same way most people do… by accident.

Procurement sat at the intersection of everything that interested me. It was big picture and detail-oriented… all at the same time. It required understanding contracts and risk, but it also required thinking like a business owner. It was everything I loved about running my own firm and everything I liked about practicing law without that “doing homework for a living” feeling I hated so much.

Every decision had consequences. You had to understand not just what was written on paper, but how it would play out in the real world. You had to understand the business, its priorities, its risks, and its goals.

It wasn’t theoretical. It was practical. It was real.

Procurement forces you to learn how a business actually operates. You see where money goes. You see what the company truly depends on. You see which systems keep things running and which partners are critical to success and how to make them better. You gain visibility into every function… finance, operations, technology, legal, leadership… because every one of them relies on external partners to do their jobs.

Very few roles give you that kind of perspective.

You begin to understand not just what the company does, but how it works. You see how decisions are made, what drives them, and what the downstream impact looks like months or years later. You learn how to evaluate tradeoffs. You learn how to balance cost, quality, speed, and risk. You get to use that law school skill of asking the right questions, because the right question often matters more than the right answer.

Law School taught me to think a certain way. Procurement teaches you how to communicate that thought.

Procurement requires you to work with stakeholders who are experts in their fields but may not be thinking about commercial structure or long-term risk. It requires you to work with suppliers whose job is to present their solution as the only viable option. It requires you to translate business needs into contracts, and contracts into business outcomes. You learn how to build trust. You learn how to influence decisions. You learn how to bring clarity to situations where there isn’t always an obvious answer.

Over time, you develop judgment.

You learn to recognize patterns. You learn which risks matter and which ones don’t. You learn when to push and when to move forward. You learn how to protect the organization without slowing it down unnecessarily. You learn how to think not just about the decision in front of you, but about the long-term health of the business.

Those skills stay with you forever.

One of the things that makes procurement such a great career is that it teaches you how businesses really function. Not the org chart version. Not the polished presentation version. The real version. The version where decisions have tradeoffs and consequences. The version where relationships matter. The version where good judgment creates real value.

It also builds skills that apply everywhere. Every company needs procurement. Every industry relies on people who understand how to evaluate partners, manage risk, and make sound commercial decisions. Once you learn how to do it, you carry that knowledge with you. It compounds over time. You become more effective, more confident, and more valuable with every experience.

Procurement isn’t about buying things.

It’s about helping organizations make better decisions.

It’s about protecting them from risk.

It’s about enabling them to grow.

It’s about understanding how to create value in ways most people never see.

I never planned to build a career in procurement. But once I found it, I realized how much it had to offer. It challenged me. It taught me and it still does. It gave me a perspective on business that I never would have gained anywhere else. That is exactly what I needed!

And the truth is, there are probably thousands of people out there right now who would love this career if they knew it existed.

They just haven’t found it yet because it has done a horrible job of marketing itself.

So, check your local high school for a career fair and get a table ready!

So, how did you end up in Procurement?

And, what keeps you here?

They’re wrong … but it’s ok because we are all wrong sometimes.

That magic, its not in the boardroom strategy decks or in the glossy investor presentations. It wasn’t developed by the Big 4 consultant that your Chief Transformation used to work for and brought in to change the future for the company. It doesn’t happen at the expensive offsite C-level meetings. It’s not the overly fabricated all-hands or the cutesy newsletter you just deleted from your inbox.

That magic happens in a few cubicles in the back where people don’t go unless they have to for some sad reason. The dark space after the pride lands…

That’s where your sourcing and contracts team live and make that magic.

The magic is in the contracts, negotiations, supplier decisions, and quiet moments where someone decides whether the company saves money, manages risk, or creates leverage… or gives it away.

That moment is The Procurement Experience and this is a real-world look inside.

This isn’t theoretical babble. It is battle-tested experience and expertise at work! (You’re going to hear this a lot over our journey together.)

Why is this starting? Well, Procurement blogs and podcasts are so boring. Everyone is so proper and polished. It’s not real. It’s not the life my friends and I live in the trenches. That’s what we should be talking about…

Here, you’ll find insights from the front lines like how deals actually get done, where companies leave money on the table, how suppliers think, and how procurement, when done right, becomes one of the most powerful drivers of profitability and operational control in any organization.

We’ll cover (this list is just the tip of the iceberg):

• The reality of procurement as a career and why it’s one of the most misunderstood and valuable functions in business

• Negotiation tactics, contract strategy, and supplier management that create real leverage

• Commentary on industry trends, vendor behavior, and procurement technology

• Practical breakdowns of deals, mistakes, wins, and lessons learned

• And yes, why you should engage with us… because Procurement Counsel is a business after all

Whether you’re a procurement professional, a business owner or team member responsible for vendor spend, a founder trying to scale efficiently, or just someone curious how businesses actually work behind the scenes, this publication is for you.

Procurement isn’t paperwork. It’s power.

It’s the power of negotiation, reading between the lines, understanding people, priorities, and the evolving landscape of the business… all at the same time. It’s keeping the fires contained and stakeholders happy while keeping the business safe.

I once heard someone say ADHD was the key to being a good procurement professional… and that hit me hard. This is why we stay unmedicated because the key to making this magic is being able to do 12 things at once and balance countless shifting priorities all while dodging grenades.

And once you see how it really works, you’ll never look at business the same way again.

This is Procurement Unmedicated.

Let’s go on a journey together, shall we?