Why Your Training Scores Are Lying to You (And How to Measure What Actually Changes)

Ananya had great numbers.

She was the L&D head at a 600-person BFSI company in Pune. Every quarter, she presented her training report to the leadership team. Satisfaction scores: 91%. Attendance: 96%. Trainer ratings: 4.6 out of 5.

She felt good about these numbers. Her CFO did not.

“Ananya,” he said at the Q3 review, “we spent ₹28 lakh on leadership training this year. What changed?”

She had no answer.

This is the training ROI problem playing out across India every quarter, and most L&D teams are not equipped to answer it.

Not because the training had been badly designed or poorly delivered. But because she had never been asked to measure anything beyond whether people enjoyed the programme.

This is the training ROI problem in India. And it is far more common than most L&D professionals are willing to admit.

Training ROI in India is the measurable business return on money spent on corporate training, proven not by satisfaction scores but by evidence that behaviour changed on the job and a defined business metric moved.

The Training ROI Illusion India Cannot Afford

Walk into any corporate training debrief in India, and you will find some version of the same ritual. The session ends. The facilitator hands out feedback forms, sometimes called “smile sheets” in the industry, and for good reason. Participants rate the trainer, the content, the venue, and the food. The scores are compiled. The report goes to HR. The report goes to the CFO. Everyone nods.

Nobody asks what changed on Monday morning.

This is Level 1 of the Kirkpatrick Model Reaction, and most Indian organisations treat it as the finish line. It is not even close to the finish line. It is the starting block.

Donald Kirkpatrick developed his four-level training evaluation framework in 1959. Over six decades later, it remains the most practical, most widely validated model for measuring whether training actually works. And in our work at Excellential across startups, mid-sized companies, and large enterprises in India, we find the same pattern consistently: organisations measure Level 1 religiously and Levels 2, 3, and 4 rarely.

That gap is where training budgets disappear.

The Four Levels and What Indian Organisations Actually Do with Them

The training ROI conversation in India almost always stops at Level 1.

Level 1: Reaction Did participants find the training useful, relevant, and engaging?

This is what your end-of-session feedback form measures. It matters that a training that nobody engages with is unlikely to change behaviour, but satisfaction scores tell you almost nothing about whether learning happened or behaviour changed. A stand-up comedian can get a 5 out of 5 on a feedback form. That does not make the session developmental.

Most Indian companies measure this.

Level 2: Learning Did participants actually learn what the training intended to teach?

This means testing for knowledge gain, skill acquisition, or attitude shift. Pre-and-post assessments. Role-play evaluations. Knowledge checks at the end of Day 1 and again three weeks later to see what was retained.

Most Indian companies: skip this entirely.

The reason? It feels like extra work. It also makes the training more accountable, and accountability cuts both ways. If the assessment shows participants did not learn, that is a harder conversation than a 4.7 satisfaction score.

Where Most Indian Organisations Actually Fall Short

Level 3: Behaviour. Are participants applying what they learned back on the job?

This is the level that separates training that changes organisations from training that entertains people for two days. It requires observation, manager feedback, and a 60–90-day window after the programme ends.

A manager who attended a feedback skills workshop: Is she actually having different conversations with her team? A sales executive who went through negotiation training: Is his close rate changing? A first-time manager who completed a delegation programme: Is he still doing everything himself, or has something shifted?

Most Indian companies do not track this at all.

The reason? No one owns it. The L&D team delivered the training and moved on. The line manager was not part of the programme design and does not know what to look for. Nobody set a baseline before the training, so there is nothing to compare.

Level 4: Results. Has the training produced measurable business outcomes?

This is the level the CFO cares about. Revenue, retention, CSAT scores, error rates, promotion readiness, and time-to-productivity for new hires. The business outcomes the training was designed to influence.

Most Indian companies have never connected a training programme to a business outcome.

And this is where Ananya found herself, sitting in front of her CFO with no answer.

Why We Stop at Level 1

In our experience working with L&D teams across India, three things cause organisations to stop at reaction data.

First, nobody defines success before the training begins. The L&D team designs the programme, books the facilitator, and coordinates logistics. But they rarely sit with the business leader and ask: “In 90 days, what should be different? How will we know this worked?” Without that conversation, there is no measurement framework, only a feedback form.

Second, L&D is measured on activity, not impact. In most Indian companies, the L&D function is evaluated on training hours delivered, the number of programmes run, and attendance percentages. Not on business outcomes. When your KPIs do not require you to measure ROI, you do not measure ROI.

Third, fear of accountability. If you measure Level 3 and find that behaviour has not changed, you have to answer an uncomfortable question: was it the design, the facilitator, the follow-up, or the organisation that failed? Smile sheets never force that question. Level 3 and 4 data always will.

What Measuring Training ROI in India Actually Looks Like

It does not require expensive software or a research team. It requires intention and a plan built before the training starts.

Before the program: Sit with the business leader and define two things: the behaviour you want to change and the business metric that behaviour affects. If you are running a conflict resolution programme for a 150-person operations team in Chennai, the behaviour change might be “managers escalate fewer issues to HR,” and the metric might be “HR escalation rates drop by 30% in six months.” Write that down. That is your Level 4 target.

During the programme: Run a pre-assessment. Even a simple one, such as a scenario-based questionnaire, a role-play observed by the facilitator, or a self-rating scale. This is your Level 2 baseline. Run the post-assessment on the last day. You now have a before-and-after on learning.

30 days after: Send a short follow-up survey to participants and their managers. Not a feedback form. A behaviour observation checklist: “In the past 30 days, have you seen this person apply [specific skill] on the job?” This is your Level 3 data.

90 days after: Pull the business metric you defined before the programme began. Compare. That is your Level 4 data. It will not always show a clean causal link; business outcomes are influenced by many factors. But consistent movement in the right direction, alongside strong Level 3 evidence, is a defensible story for any CFO.

A Real Example

We ran a leadership development programme for a life insurance company in the APAC region 25 senior managers, focused on delegation and emotional intelligence. Before the programme, delegation success rates across the group sat at 35%. Managers worked over 12 hours daily. No new initiatives had been launched internally in 18 months.

• We started with a diagnostic.
• We defined the Level 4 target: raise delegation success rates and restart the internal innovation pipeline.
• We built Level 2 assessments into the programme design.
• We ran 1:1 coaching follow-up at the 60-day mark to track Level 3 behaviour change.

Three months after the programme ended, delegation success rates had moved from 35% to 92%. Eight innovation ideas per month were being generated by teams that had previously been silent. The business had a measurable answer to “what changed?”

That is what training ROI looks like when you build the measurement into the programme from Day 1, not after it ends.

The Question to Ask Before Your Next Training Budget Gets Approved

Before you sign off on the next programme, ask three questions:

1. What specific behaviour are we trying to change?
2. How will we know in 90 days whether it changed?
3. What business outcome will we track?

If you cannot answer all three, the training is not ready to run. The design needs to go back to the drawing board, not because the content is wrong, but because, without measurement, there is no ROI. There is only activity.

India Inc. spends thousands of crores on corporate training every year. The organisations that get results from that spending are not the ones with the biggest budgets. They are the ones who decided, before a single slide was designed, exactly what success would look like.

At Excellential, we design every leadership and behavioural training programme with the Kirkpatrick framework built in from the first planning conversation. We help L&D teams define success at Level 4, design for Level 3 behaviour change, and build the measurement systems that give leadership teams a real answer when they ask: what did it change?

At Excellential, we treat training ROI in India as a design question, not an afterthought.

Frequently Asked Questions

What is training ROI?

Training ROI is the measurable business return on money spent on training, set against its cost. It goes beyond satisfaction scores to ask whether participants learned, changed their behaviour on the job, and moved a business metric the training was designed to influence. Real ROI is proven, not assumed.

What is the Kirkpatrick Model?

The Kirkpatrick Model is a four-level framework for evaluating training, developed by Donald Kirkpatrick in 1959. The levels are Reaction (did they like it), Learning (did they learn), Behaviour (are they applying it), and Results (did business outcomes move). Each level builds on the one before.

Why do most Indian companies fail to measure training ROI?

Most stop at Level 1 because no one defines success before the programme, L&D is measured on activity rather than impact, and Level 3 and 4 data invites hard accountability. Without a baseline or an owner for follow-up, there is nothing to compare and no ROI to report.

How do you measure training ROI in India?

Before the programme, define the target behaviour and the business metric it affects. Run a pre and post assessment for learning, a 30-day behaviour check with managers, and a 90-day comparison of the business metric. This gives you Level 2, 3, and 4 evidence without expensive software.

What is the difference between Level 3 and Level 4 in the Kirkpatrick Model?

• Level 3 measures behaviour: whether people apply the training on the job, tracked through observation and manager feedback over 60 to 90 days.
• Level 4 measures results: whether that behaviour moved a business outcome such as retention, CSAT, or close rate.
• Level 3 explains Level 4.

Write to us at support@excellential.com if your training budget deserves a better answer than a smile sheet. Book a Free Consultation