The 90-Day Test: How to Know If Training Actually Worked

Ask most organisations how their last training programme performed and you'll get attendance figures, satisfaction scores, and a facilitator rating. Ask what changed in the business as a result, and the room tends to go quiet.

That's not because the training failed. It's because it was never measured against the right question. Completion tells you who showed up. It tells you nothing about whether anyone does their job differently as a result.

MEASURING AT THE WRONG MOMENT

Most evaluation happens at the end of the session, when enthusiasm is highest and behaviour hasn't had a chance to either take hold or fade. That's the wrong moment to measure. The honest test of a learning investment is not how people felt in the room. It's what they are doing 30, 60, and 90 days later.

This window matters because it is where the difference between excitement and habit becomes visible. Two-week follow-ups capture enthusiasm. Ninety-day follow-ups capture whether anything actually changed.

THREE STRUCTURAL FIXES

Take a baseline before the programme starts.

You cannot demonstrate a change you never measured the starting point for. "The team responded well" is not evidence. "Scores rose from 2.3 to 4.1 across 40 managers over 90 days" is.

Track one behaviour, not everything.

Programmes that try to shift five behaviours at once rarely move any of them convincingly. Identify the single highest-value behaviour the programme is designed to change, and measure that one thing rigorously over the 90-day window.

Report what didn't work, not just what did.

The reports that get trusted in a budget meeting are the ones that admit where a number didn't move. A report claiming universal success reads like marketing. A report naming two out of five behaviours that improved — and being specific about the three that didn't — reads like evidence, and evidence is what earns the next budget approval.

THE LINK TO A BUSINESS METRIC

Skill improvement is a means, not an end. The strongest evaluations draw a straight, honest line from the behaviour that changed to something the business already tracks — retention, error rate, time to competency, conversion. Without that link, even a genuine improvement is hard to defend when budgets get reviewed.

The 90-day test is simple to state and rarely applied: if you can't demonstrate that behaviour changed, you can't claim the investment worked. Before your next programme launches, it's worth deciding now what you'll be able to say about it in 90 days.