The Founding Five: Five firms. We build the follow-through on your program over summer (winter in the Northern Hemisphere), so you're selling it to your clients from February. The Founding Five: built over the break, selling to clients from Feb How it works Merry Christmas from Australia & Japan · 良いお年を

October 31, 2026

How to Measure Leadership Training Effectiveness on the Job

How to measure leadership training effectiveness beyond smile sheets: track behaviour at 30, 60 and 90 days and give buyers proof they can take upstairs.

The short answer

Measure what leaders do differently on the job, not how they felt on the day. Track two or three behaviours from your model at 30, 60 and 90 days, using a short self-report plus one manager observation per person, and show them next to a couple of numbers the buyer already tracks. Promise behaviour evidence, not ROI. The easiest way to collect that evidence is a follow-through phase you run between and after sessions.

Why is leadership training so hard to measure?

Because most of the measuring happens in the room, and most of the change is meant to happen outside it. In the LEADx 2024 benchmark, about 90% of organisations measured reaction, only 39% measured behaviour change and 22% measured business impact. LEADx is a vendor survey of 145 companies, and LEADx sells reinforcement, so treat the numbers as a signal, not a census. Fosway’s 2025 research found barely 2% of L&D professionals say they’re advanced at measuring impact.

So the gap isn’t a lack of evaluation models. It’s that nobody stays in contact with participants after the last session, so there’s nothing on the job to measure.

For a leadership or L&D firm, that shows up as three problems.

Problem 1: you only get paid for the event

Your fee usually ends when the program does. Measuring behaviour at 60 and 90 days is work that happens after the invoice, so it rarely gets done. If measurement isn’t part of what the client pays for, it stays a good intention in the proposal.

The fix (Sell it): make the follow-through, and the measurement inside it, ongoing support the client pays for every month.

Problem 2: it isn’t used on the job, and nobody can prove it worked

The same LEADx benchmark found 75% of leadership development leads estimate that less than half of training is applied. Harvard Business Review’s Why Leadership Training Fails, and What to Do About It (Beer, Finnström and Schrader, 2016) explains why: the organisation’s systems pull people back to the old way of working. Without evidence of behaviour change, your buyer can’t defend the budget.

The fix (Prove it): a scorecard of what people actually do differently, filled in by hand at first.

Problem 3: no one has time to follow through

In Training magazine’s 2025 Training Industry Report (US), 41% of training teams said a lack of resources or people was their biggest challenge. The client’s team won’t chase 30-day check-ins.

The fix (Grow it): follow-through that runs for them every month, so the evidence keeps arriving without their team doing the work.

What should you measure? Kirkpatrick’s four levels, used honestly

Kirkpatrick’s model gives you four levels:

  1. Reaction: did people find it useful and relevant?
  2. Learning: do they understand the model, and can they show it?
  3. Behaviour: are they doing it on the job?
  4. Results: did anything the business cares about move?

Most firms stop at level one or two. The useful shift is to put most of your effort into level three and treat level four with care. Business results depend on many things besides your program. Behaviour you can observe and record yourself.

How do you measure behaviour change after leadership training?

Keep it small and specific:

  1. Pick two or three behaviours. Take them straight from your model. “Runs a weekly one-on-one with each direct report” beats “communicates better”.
  2. Set a baseline. Before the program ends, ask participants and their managers how often each behaviour happens now.
  3. Check at 30, 60 and 90 days. A short self-report from the participant plus one observation from their manager. Five questions at most.
  4. Capture examples in their words. One real situation per check-in where they used the behaviour. Buyers remember stories more than scores.
  5. Report on one page. The behaviours, the trend, then two or three examples. That’s the scorecard your buyer can take upstairs.

What about leadership development ROI?

Be careful here. A dollar ROI figure for leadership training usually rests on assumptions a sceptical finance team will pull apart. It’s more credible to show behaviour evidence first, then point to two or three numbers the buyer already tracks for that team, such as retention, engagement scores or time to resolve issues. Don’t claim the program caused them to move. Show the behaviour change, show the numbers alongside it, and let the buyer draw the link.

How do you collect the evidence without extra work?

Make measurement part of the follow-through, not a separate survey. Each check-in already asks people to use one part of your model on real work that week. Recording what they did is the measurement, and the manager prompt is the observation. Run it by hand for your first one or two clients, with a shared sheet and short emails. Don’t build anything until a client has paid for it and you’ve seen it work.

The 3-step fix for L&D firms and coaches

  1. Sell it. Offer the follow-through, scorecard included, to clients who just finished a program with you, as ongoing support they pay for every month. Sell it before you build anything.
  2. Prove it. Run the check-ins, practice and manager prompts between sessions yourself, and keep the behaviour scorecard by hand.
  3. Grow it. Once it works, it runs for them every month, and the evidence keeps coming.

Measurement stops being a cost you absorb and becomes part of what clients pay for.

Frequently asked questions

What is the best way to measure leadership training effectiveness?

Measure behaviour on the job. Track two or three specific behaviours from the program at 30, 60 and 90 days, using a self-report plus a manager observation, and report the trend on one page.

Is the Kirkpatrick model still useful for leadership programs?

Yes, as a structure. Its four levels (reaction, learning, behaviour, results) are a reminder that reaction scores are only the start. Put most of your effort into behaviour.

How soon after training should you measure behaviour change?

Take a baseline before the program ends, then check at 30, 60 and 90 days. Ninety days is long enough to see new habits and short enough for a buyer to say yes to a pilot.

Can you calculate ROI for leadership training?

You can, but the figure usually depends on assumptions buyers don’t trust. Behaviour evidence shown next to numbers the buyer already tracks is more credible.

Do I need software to measure behaviour change?

No. A shared sheet, short emails and a manager prompt are enough for your first clients. Build only what you’ve proven clients pay for.

Take the Follow-Through Check · 2027 Edition

Describe one of your programs and see what the follow-through could look like, including how you’d prove what changed.

Take the free Follow-Through Check

The Founding Five. Five firms. We build the follow-through on your program over summer (winter in the Northern Hemisphere), so you're selling it to your clients from February. How it works

Related

About the author

Matt Ainsworth is the founder of The Lean CPTO. He has 20+ years across product, technology, sales, marketing and communications in Australia and nearly a decade in Japan, has mentored founders for 10+ years, and helps leadership, L&D and consulting firms turn a program or method into ongoing support clients pay for every month.

Start with one program.

Describe one program and the free Follow-Through Check shows you the follow-through that keeps it working on the job, a sample session on your own model, and what your client would see at renewal, in about 2 minutes.

The Product Pick call is free and takes 30 minutes. We pick the program with the best shot and map how you'd pre-sell it to 3 clients. You leave with that plan either way.