We Answered All 37 ESOMAR Questions. Here’s What Most Providers Won’t.

This post opens a short series unpacking what’s actually in our ESOMAR 37 responses — because a compliance document buried on a website doesn’t help anyone make a better buying decision.

---

There’s a document almost nobody asks their sample provider for, and almost every provider hopes it stays that way.

It’s called ESOMAR 37 — a standard set of questions published by ESOMAR to help buyers evaluate the sample they’re purchasing: where it comes from, how it’s verified, what happens when things go wrong, and what a provider will and won’t disclose along the way. It exists precisely because “trust us” isn’t a methodology.

We answer all 37. Not the easy dozen. All of them, including the ones that are mildly uncomfortable to answer honestly — like what we do when a project can’t be completed in field, or how much of our sample is genuinely proprietary versus supplemented from third parties.

This post is the start of a short series unpacking what’s actually in our responses, because a compliance document buried on a website doesn’t help anyone make a better buying decision. A blog post might.

Why most providers don’t do this

It’s not that ESOMAR 37 is a secret. It’s that fully answering it requires saying things out loud that are easier to leave vague:

• Exactly what percentage of your sample is proprietary versus third-party supplemented
• What your fraud detection stack actually consists of, not just that one exists
• Whether you run a survey router, and if not, what you do instead
• What happens — specifically — when a project hits a wall mid-field

Vague answers aren’t necessarily dishonest. But they’re not useful to a buyer trying to compare providers, either. And in a market where “quality” has become a word every provider claims and few can specify, vagueness is doing a lot of quiet work.

What we’re covering in this series

Over the next several weeks, we’re pulling four themes out of our ESOMAR 37 responses and giving each one the space a compliance document can’t:

How sample actually gets built. The difference between deterministic and probabilistic sampling, and why we don’t run a traditional survey router — not because it’s a limitation, but because of what it trades away.

What buyers should be asking. The specific metrics ESOMAR recommends requesting from any provider, and why “proprietary” and “exclusive” mean two very different things that get used interchangeably.

The trust stack. What’s actually happening in the milliseconds between someone clicking a survey link and reaching question one — identity verification, device fingerprinting, behavioral analysis, and what happens to the people who don’t pass.

What happens when things go sideways. Feasibility shortfalls, source drift in tracking studies, and the difference between a provider who discloses a problem early and one who hopes you don’t notice.

The honest version of “trust us”

None of this is really about ESOMAR 37 as a checklist. It’s about whether a provider is willing to be specific when specificity is inconvenient. We think that’s a fair thing to be judged on, and a fair thing to ask of anyone selling you sample.

Next up: the questions you should be putting to every provider you evaluate — including us.