Your Training Program's Report Card: A Kirkpatrick-Based Self-Audit

Many training teams feel their programme is working, but can't point to evidence besides attendance figures and satisfaction scores. The Kirkpatrick Model offers a structured way to measure success and doubles as a useful audit. Run your training through all four levels honestly to quickly find where the gaps are.

Score yourself honestly on each level below. Most training programmes are strong at the bottom of the pyramid and thin at the top; this is normal, and it highlights the gap worth closing next.

Level 1: Reaction. Did learners find the training relevant and engaging?

Ask yourself:

Do we collect feedback after every session, or only sometimes?
Do we track trends over time, or just look at each score in isolation?
Do we act on low scores, or just note them?

If you're weak here: this is the cheapest level to fix. A short, consistent post-session survey with one open-text question is enough to get a real baseline.

Level 2: Learning. How did knowledge or skills change?

Ask yourself:

Do we ever test understanding before and after a session, even informally?
Can we say, with evidence, that a specific session improved comprehension, or are we assuming it did because attendance was good?
Is there a meaningful difference between "we covered it" and "they learned it" in how we plan sessions?

If you're weak here: add a lightweight pre/post check, even a single confidence-rating question, to your next few sessions and compare the shift.

Level 3: Behaviour. Are people applying new skills on the job?

Ask yourself:

Do we know whether attendees actually use what was taught, or do we lose visibility the moment the session ends?
Is there any follow-up mechanism, like a check-in, usage report, or manager conversation, after training happens?
If usage data exists in the product, is anyone connecting it back to who attended which session?

If you're weak here: this is usually the biggest and most valuable gap to close. Pick one taught workflow you can observe in-product, and track adoption for attendees compared to non-attendees for a month.

Level 4: Results. Did the training achieve business outcomes?

Ask yourself:

Can we point to a business metric that moved because of a specific training initiative, such as support ticket volume, adoption rate, retention, or expansion?
Does leadership see training as a cost centre or a measurable driver of outcomes, and is that perception backed by data either way?
If the training budget were cut in half tomorrow, could we make a data-backed case for what would be lost?

If you're weak here: don't try to prove this level in isolation; it's built on Level 3 data. Once you have a behaviour change you can point to, connect it to the nearest downstream business metric and let that be your first Level 4 case study, even if it's narrow.

Reading your results

A programme that is strong at Levels 1 and 2 but weak at 3 and 4 is common. This usually means the training content is sound, but you lack visibility into what happens after the session. A programme that is weak across the board needs to start at Level 1 and build up sequentially; skipping ahead to Level 4 without the proper instrumentation underneath just produces numbers nobody can defend under scrutiny.

The purpose of this audit is to honestly identify which level your evidence currently supports. This is important because "we ran great sessions" and "we can prove our sessions worked" are two very different claims, and only the latter holds up in a budget conversation.