What to ask before you buy an employee engagement survey.

That gap is worth finding before you sign a contract, not after. A vendor demo will show you dashboards, benchmarks and a clean UI. It will not show you the question that actually matters: what happens when the scores come back high and turnover climbs anyway.

Here are nine questions worth asking before you buy.

1. Does the survey measure sentiment, capacity, or both?

Most tools on the market — Gallup’s Q12, Culture Amp, Qualtrics EmployeeXM — measure engagement well. Engagement is whether the organisation motivates and enables contribution: purpose, leadership, connection, growth. It is real, and it is worth measuring.

What it does not tell you is whether people can sustain that contribution. Capacity is the physiological and cognitive headroom a person has to do their job well — energy, recovery, focus, resilience. Ask any vendor directly: does your instrument separate these two constructs, or does it fold everything into one composite score? A single blended number hides exactly the pattern you need to see — the workforce that scores well on sentiment while running on empty.

2. What happens to teams that score well on engagement and poorly on the thing you are not measuring?

This is the question most demos deflect. Engagement and capacity move independently — a team can be highly committed and still be depleted. If the tool cannot show you that pattern, it cannot warn you about it either.

3. How many statements, and how long does it take?

Long surveys get abandoned or rushed. Ours runs 55 statements, split across capacity and engagement domains plus eNPS, in about 12 minutes. Ask what the vendor’s completion rate looks like at scale — not in their best case, in their average one.

4. Is the data anonymous by design, or anonymous by policy?

Policy can change. Design cannot be quietly reversed. Ask whether individual responses are ever stored in a form that could be traced back to one person, and at what group size results start showing — ours is a minimum of five respondents per reported score, ten for a demographic cut and thirty for open text — smaller groups simply do not get a number.

5. What is the evidence behind the model?

Ask for the validation study, not the marketing deck. Behind ours: 140-plus peer-reviewed papers and a validation study of 10,000-plus working adults across the US and Europe. If a vendor cannot point you to published, citable research behind their scoring model, the model is an opinion with a UI.

6. Does the output end in a dashboard, or in an action?

A benchmark chart tells a leader where they stand. It does not tell a manager what to do on Monday. Ask whether the platform generates specific, evidenced actions tied to what the data actually shows — and whether those actions come with a way to check, next cycle, if anything moved.

7. Who sees what?

Employees should see their own results, privately. Leaders should see aggregated, anonymised patterns — never an individual’s score. If a platform’s architecture allows an individual score to surface anywhere in a manager’s view, that is a trust problem waiting to happen, not a feature.

8. What is included beyond the survey?

A number without a next step is a report nobody reads twice. Ours pairs the measurement with Thrive Academy — short, targeted learning matched to what the results show — so the finding and the fix live in the same platform.

9. What does it cost to run this properly, not just to pilot it?

Pilot pricing hides the real number. Ask for per-employee-per-year pricing at your actual headcount, and ask what is included at that price versus what is an add-on.

The tools worth buying are the ones that will tell you, honestly, when your engagement scores are climbing and your capacity is quietly running out from underneath them.