Search for coaching ROI and you will find confident multiples — seven times the investment, five times, eleven. Trace them and most rest on self-reported estimates from participants asked to guess the financial value of their own development, in surveys commissioned by organisations selling coaching.
That does not mean coaching does not work. It means those numbers are not evidence, and repeating them to a sponsor who checks is a bad trade.
What follows is what you can honestly claim.
What You Cannot Measure
- Causal attribution. A coached manager whose team improves may have improved anyway, or because of a reorganisation, or a new hire. Without a control group you cannot separate these, and you do not have a control group.
- Financial return. Converting "communicates more clearly" into currency requires assumptions that dominate the result.
- Counterfactuals. What would have happened otherwise is unknowable, and it is where most inflated claims hide.
Say this plainly when asked. Sponsors who take measurement seriously respect it, and the ones who wanted a number to justify a decision already made will get one from someone else regardless.
What You Can Measure
1. Self-rated progress against a stated baseline
Weak as science, genuinely useful in practice — provided the baseline was captured at the start rather than reconstructed later. "Where are you on this, 1 to 10" at intake and repeated throughout gives you a trend that the client cannot retrospectively rewrite. That is the whole reason intake baselines matter.
2. Behavioural markers the client defined
Stronger, because they are observable and specified in advance. "I will have the conversation with my manager." "I stop checking email after 7pm." Whether they happened is a fact, not an impression.
3. Third-party observation
The most credible thing available to you. A short before-and-after from two or three colleagues, on the themes agreed at kickoff. Not a full 360 — four questions is enough, and it survives the erosion problem described in showing a client they have changed, because other people's baselines do not move with the client's.
4. Engagement
Attendance, completion of between-session work, check-in consistency. Not an outcome, and honest process evidence — and for an organisational sponsor, the one metric you can report without touching content.
How to Answer "What's the ROI?"
Something close to this:
"I can't give you a credible financial multiple, and I'd be suspicious of anyone who does. What I can give you is: the development themes agreed at the start, self-rated progress against them across the engagement, whether the specific behaviours we identified actually happened, and observations from the people around them. That is real evidence. It is not a number with a currency symbol."
This loses you the occasional buyer and wins you the ones worth having.
Measure Little, Consistently
The instinct when asked for evidence is to add instruments — validated scales, longer reviews, more questions. Completion collapses, and a rich measure nobody completes is worth less than a crude one they do.
Two numbers and a yes/no, weekly, for the whole engagement will beat any assessment battery administered twice. The consistency is what makes it a trend, and the trend is the only thing that constitutes evidence.
Where the Effort Actually Goes
The work in measurement is not designing it — it is collecting it every week for six months without it becoming a chore that quietly stops. Manual systems fail at exactly this point, for the reasons in tracking progress without admin.
Automating the collection is the difference between having evidence at the end of an engagement and having a recollection. That is what Interlude is for — and the reason the honest version of this answer is available to you at all.
Sign in to join the conversation.
Sign In