77% of enterprises re-evaluate AI vendors every 6 months

New research on enterprise AI buying says the quiet part out loud: 77% of enterprises re-evaluate their AI vendors at least every six months, and 29% do it on a rolling basis. That number got a fresh round of attention this week, and it is usually read as bad news for AI startups. Read from the buyer's side, enterprise AI vendor re-evaluation at that cadence is leverage — if your stack is built so you can act on it.

What actually happened

The figures come from Madrona's report Harnessing Enterprise Value: The ROI of AI, published August 13, based on a survey of 150 senior enterprise decision-makers. Three numbers carry the argument:

• 74% plan to expand AI budgets over the next 12 months.
• 83% converted fewer than half their AI pilots into production over the last 12 months.
• 77% re-evaluate AI vendors at least every six months.

Madrona ranks the reasons pilots die: integration complexity first, then security, privacy and compliance requirements, then ROI scrutiny. The report also finds outcome-based pricing is the model buyers most prefer and least often encounter, while usage-based pricing shows the exact inverse. TechCrunch, covering the research on September 3, pairs it with an a16z survey of 50 technical AI buyers and Madrona's own framing of a "fast in, fast out" dynamic that does not look like traditional enterprise SaaS.

Why vendor re-evaluation matters for your business

Start with the number that should sting: 83% converting fewer than half their pilots, with integration complexity as the top cause. That is not a model quality problem. It is a plumbing problem — the pilot ran on a spreadsheet export and a demo key, and nobody costed the work of wiring it into the systems of record. If you are running a pilot right now, the honest question is not "did the AI do well," it is "what would it take to put this in the path of real transactions," and that number should exist before the pilot starts.

Then use the re-evaluation cadence deliberately. A six-month review is only leverage if switching is actually possible in six months. That means the things that make you portable have to be yours: prompts and evals in your repo, not in a vendor console. Your data in your warehouse, exported on a schedule you control. The vendor call behind an interface you wrote, so a swap is a config change and a regression run rather than a rebuild. Teams that skip this discover their review is theater — they re-evaluate annually, conclude they are stuck, and renew.

Last, take the pricing finding seriously as a negotiating position. Buyers want to pay for outcomes and vendors keep selling usage. You will rarely get a pure outcome contract from a large vendor, but you can get closer: a shorter term, a documented success metric, and a clause that ties renewal to it. Ask. The re-evaluation cadence is why they will listen.

Sources: Madrona, TechCrunch.