How do you know whether a learning project has actually been successful? In this blog, Andy Candler, COO of L&D Free Spirits, shares five practical tips drawn from years of working with clients, from asking the right questions at the start to choosing measures that genuinely reflect the problem you’re trying to solve.
Five tips I live by in my work
Here’s a question that sounds simple but trips up almost every L&D project I’ve ever worked on:
How are we going to measure success?
I ask it right at the start of every client workshop I run. Then, right at the end of the same workshop, after an hour or two of digging into the detail of what a client actually needs, I ask it again. The answer is almost never the same the second time.
This change is so very important. It tells you everything about how much clearer the project has become, and it’s where most of what I know about measuring success actually comes from.
So here are five tips, built from years of asking that question twice and comparing the answers.
1. Set yourself up for success before you measure anything.
You can’t measure success on a project built on a shaky relationship. Sounds obvious, but it’s the bit people skip because they’re keen to get into the fun stuff like content, solutions and delivery.
Before any of that, get the project Roles and Responsibilities completed. Who owns what, who signs off, who’s accountable for pulling the data afterwards. It sounds like admin, which it absolutely is, but it’s the admin that stops the confusing questions when you have rolled up your sleeves and got into the project delivery.
A good Roles and Responsibilities document isn’t there to protect anyone from blame. It’s there so that measuring success is actually possible, because someone specific has agreed to help make it happen.
2. Ask the question twice and challenge the data.
This is the heart of my Discovery Workshop method. Ask “How will we measure success?” at the very start, before anyone’s thought too hard about it. You’ll usually get a client looking like a rabbit in the headlights, and they will mutter something vague like: “Engagement.” “Better performance.” “Change.”
And that’s absolutely fine. Don’t dwell on it, the purpose of the question was to get clients into that mindset. It’s the perfect starting point for your discussion.
Then ask it again, at the end of your workshop. By now the conversation has forced the client to think properly about what they actually need, and how we will solve it, and the second answer is usually sharper, more specific, and a lot more honest.
But don’t stop there. When you get that second answer, push back on it. Ask directly: “Does the data to measure this actually exist? Can someone pull it? Is it already being collected somewhere, or does someone need to start collecting it from scratch?”
I’ve lost count of the number of projects where the client wanted to measure something brilliant, and the data simply wasn’t there. Better to find that out in the workshop than three months after delivery when it’s too late to fix.
3. Give it time to breathe.
One of the most common mistakes is trying to measure success immediately after delivery. There simply hasn’t been enough time for anything to show up in the data yet.
As a rough guide:
- Short, finite projects (a single course, a defined skills rollout, something with a short deadline) – three months is usually sufficient time to start seeing something meaningful in the results.
- Longer-term or embedded projects – six months is the minimum, and I wouldn’t go beyond twelve months before you’re looking for a first read on impact.
Measure too early and you’ll see noise, not real indications of success. Measure too late and the business has moved on, and nobody’s interested in last year’s project anymore. Getting that window right matters as much as picking the right metric in the first place.
4. Look beyond ROI. ROE matters too.
There’s often a lot of noise in L&D about Return on Investment (ROI), and whether every learning project needs to prove a financial payback. Here’s a more useful way to think about it: not every project has an immediate, obvious financial impact, and that’s absolutely fine.
Sometimes the right measure is Return on Engagement (ROE). Did the right number of people complete the learning in the time they needed to? Did they hit the pass marks? Did they actually show up and take part?
That’s a legitimate success metric in its own right, especially for compliance training, onboarding, or anything where the real business value is in reduced risk rather than increased revenue.
Don’t let the ROI conversation talk you out of a measure that’s genuinely meaningful, just because it doesn’t have a pound sign attached to it.
5. Think outside the box. Or choose another box.
This is where I think things get interesting. The most obvious metric isn’t always the correct one to be concentrating on.
Let me start with a real-world example from a retail client I worked with:
We were training store colleagues on how to sell products properly, and the instinctive metric would have been additional units sold. More stuff going out of the door.
But the real business problem was product returns. Customers were buying products, getting them home, and finding they didn’t fit their lifestyle or their existing setup, then bringing them straight back. So, the success metric for that project wasn’t sales at all. It was how many products came back through those same doors a few days, sometimes hours, after.
A drop in returns told us the colleagues were actually listening to customers and recommending the right product, not just pushing for a quick sale and more commission.
Obviously, metrics will be very specific to your; project, identified problem, and learner needs. But I wanted to give you some thoughts around how to think outside-the-box.
Some ideas worth having in your back pocket when you are talking to clients:
- Time to competency – how long it takes someone to work unsupervised after training, rather than just whether they passed the course.
- Support ticket or help desk volume – a genuine drop after a process or systems training rollout is a strong, tangible signal.
- Error or rework rates – particularly useful for technical or compliance-heavy training.
- Retention – are people who went through a strong onboarding or development programme staying longer than those who didn’t?
- Manager pulse checks – a quick, informal check-in at 30, 60 and 90 days, asking managers whether they’ve actually noticed a behaviour change. Simple, cheap, and often more honest than a survey score.
- Staff happiness – a simple before-and-after read on morale or job satisfaction. It won’t always show up on a balance sheet, but happier, more confident staff tend to feed into the other numbers anyway, lower turnover, fewer errors, better customer interactions, so it’s often an early signal worth tracking properly.
None of these examples will fit every project. That’s rather the point. The skill isn’t finding one universal success metric, it’s asking the right questions, twice, and being willing to follow the answer wherever it actually leads.
The takeaway
Measuring success isn’t something you bolt on at the end of a project. It starts off on the right foot by identifying who works on the project and what they will be doing to make it a success. It then progresses to the very first measurement conversation, gets properly defined by the second, and only means anything if you’ve built in the time and the data to prove it.
Get this right, and it does something important for you too. It shows your client you’re operating at a genuinely professional level, someone who looks at the relationship holistically rather than just delivering and moving on. That’s exactly the kind of reassurance that makes clients want to work with you again and again.
