Buy judgement, not hours: commissioning research for a changing market

If agencies are under pressure to change, so are the teams that buy from them. Greenbook’s GRIT research found that only 34% of insight buyers had backing for bigger budgets, down from 65%, while the same buyers are asking for analytics depth, operational integration and machine-augmented speed. More, faster and better, on the same money. Meanwhile ESOMAR finds clients now run 48% of projects in-house, yet only 4% report a strategic AI rollout. The client-side team is becoming an orchestrator of evidence, whether or not anyone has updated its job description.

From buying hours to buying outcomes

Most research is still commissioned and priced as a quantity of effort: interviews, sample, days, slides. In a market where AI can compress much of that effort, the useful questions to ask a supplier have changed:

• ·Who will do the work? The people in the pitch should be the people on the project.
• ·What decision is this for? A study without a named decision is a cost, not an investment.
• ·Where is AI used, and how is it checked? Speed is welcome; unvalidated speed is a liability.
• ·What does the fee actually buy? Thinking, or the overhead that surrounds it?

Fit for purpose, hybrid by design

There is no single best method, only the one that fits the decision. Three tests do most of the work: the stakes of being wrong, how reversible the decision is, and how novel the behaviour being studied. Low-stakes, reversible and familiar questions suit fast, AI-led approaches. High-stakes, irreversible and novel ones call for real people, robust samples and validated measures. Most commercial questions sit in between, which is why the strongest programmes are deliberately hybrid.

Purpose before process

Above all, research earns its budget when it changes a decision. That means starting from the commercial question, measuring what people actually do rather than what they say they will do, and ending with an action someone owns and a metric that shows whether it worked. A useful house rule: no study commissioned without a named decision, and no debrief closed without a named action.

None of this needs a bigger budget. More often it needs smaller, sharper studies with experienced people involved from the first conversation. If you have a question taking shape and would value an independent perspective on how to approach it, we are always glad to help think it through.

Sources: ESOMAR, Global Market Research 2025; Global Prices Study 2025; Global Users & Buyers 2025 (via Research World). Greenbook, GRIT Business & Innovation 2025 and GRIT Insights Practice 2026.