Coaching works.
Now prove it.
Measure competencies objectively before the first session. Measure them again at 90 and 180 days. The difference between those two readings is the only coaching evidence a finance team has ever accepted.
Key facts
- 01Every serious coaching-ROI framework published in 2025–26 gives the same first instruction: establish a baseline before the engagement starts.
- 02The ICF's own guidance still leans on coachee self-ratings — the weakest available instrument, and the one most easily dismissed by a buyer.
- 03Practical cadence for coaching: baseline in the first two weeks, re-measure at 90 days (early indicator) and 180 days (the result).
- 04innerly.me scores 17 behavioural indicators across four competencies algorithmically — no human rater, so the same person is judged against the same standard both times.
- 05A competency delta is not a financial ROI figure. Anyone selling you the second one from the first is selling you a correlation as a cause.
The evidence problem nobody in coaching wants to name
Coaching has a genuine effect. Practitioners know it, coachees describe it vividly, and the qualitative evidence is abundant. What coaching does not have is a way of showing that effect to someone who was not in the room — which is precisely the person who signs off the renewal.
So the industry reaches for what it has: satisfaction scores, testimonials, goal-attainment scales the coachee fills in themselves, and — remarkably often — the coachee's own rating of their capability on a scale of one to ten, before and after. That last method is recommended in mainstream coaching guidance. It is also the method a numerate CFO will dismantle in under a minute, because a self-rating after months of coaching is contaminated by the coaching itself.
The result is a strange stalemate. Coaching budgets get cut not because the coaching failed but because nobody could show that it didn't.
Self-ratings move for the wrong reasons
The coachee has new vocabulary, a relationship with the coach, and a clear sense of what answer the programme wants. All three shift the number without shifting the behaviour.
360s drift between cycles
Rater panels turn over, standards recalibrate, relationships colour the ratings. The delta mixes real change with noise you cannot see in the output.
Business KPIs cannot be attributed
Retention improved. So did six other things during the same six months. Useful as corroboration, hopeless as primary evidence of a coaching effect.
The coaching measurement cadence
Shorter than a training programme, because the intervention is spread over months rather than delivered in a block.
Baseline, before any coaching content
The coachee completes the assessment before the first substantive session. This is the step that cannot be recovered — once coaching has begun, the starting point exists only in memory, and memory after coaching is not neutral.
Coach with the map in hand
The baseline is not just an evaluation artefact — it is a coaching input. It tells the coach which competencies are genuinely weak rather than which ones the coachee is most comfortable discussing. G&T's company package includes a results discussion with a certified coach for exactly this reason.
Interim read — is anything moving?
An early indicator, and it should be reported as one. A flat 90-day read is not failure; it is information about dosage and fit, delivered while there is still time to act on it.
Re-measure with a parallel form
Same competency structure and difficulty calibration, different scenarios. This is the number that goes on the slide: where the cohort started, where it ended, on the same instrument, over a stated window — including the competencies that did not move.
If you sell coaching, this is a sales asset
Coaches and coaching platforms rarely lose deals because the buyer doubts that coaching works. They lose because the buyer cannot take “it works” to their own budget meeting. Meanwhile the ROI research the industry leans on is old and one-dimensional, and buyers have noticed.
A coach who arrives with a baseline instrument and a re-measure plan is proposing something structurally different from a coach who arrives with testimonials. It also changes the conversation about price: you are no longer asking for trust, you are offering a measurement.
Independent coaches
Add an objective diagnostic to your proposal. Baseline your client at the start, re-measure at 180 days, and hand them something they can defend internally when the renewal comes up.
Coaching platforms
Your clients face the same CFO question you do. innerly.me works as a white-label pre/post diagnostic layer underneath a coaching platform — one instrument, applied consistently across your whole book.
The underlying method is the same one we use for training programmes — read the full guide to measuring development impact, or see how a capability baseline works.
Frequently asked questions
How do you measure the effectiveness of executive coaching?
The only measurement that survives scrutiny is a comparison between two points on the same instrument: an objective competency reading taken before the first coaching session, and the same reading taken again after the engagement. Everything else in common use — post-programme satisfaction, goal-attainment scaling, the coachee's own 1-to-10 self-rating that the ICF itself suggests — measures how the coaching felt rather than what it changed. Those are worth collecting, but they are not evidence of effect, and a buyer under budget pressure will treat them accordingly.
When should the coaching baseline be taken?
Before the first session — ideally in the first two weeks of the engagement, and always before any coaching content has been delivered. This is the single step that cannot be recovered later. Once coaching has begun, the starting point can only be reconstructed from memory and self-report, both of which are systematically distorted by the experience of having been coached. A baseline taken after session one is not a baseline.
How long after coaching should you re-measure?
Coaching engagements tolerate a shorter cadence than classroom training because the intervention is spread over months rather than delivered in a block. A 90-day read is a defensible early indicator, and a 180-day read is the result. Report them as what they are: the 90-day figure shows whether anything is moving, the 180-day figure shows whether it stuck. Measuring in the week the engagement ends captures the warmth of the final session, which is real but is not a capability.
Isn't a coachee's self-rating enough?
It is the most common method and the weakest one. A self-rating moves after coaching for at least three reasons that have nothing to do with capability: the coachee has learned the vocabulary to describe themselves more precisely, they have an interpersonal relationship with the coach whose work they are implicitly rating, and they know what answer the programme is hoping for. Self-report is useful for tracking the coachee's confidence and engagement. It is not usable as evidence to a finance team, and everyone in that meeting knows it.
Can coaches use this to win business?
Yes, and increasingly they have to. Coaches and coaching platforms lose deals not because buyers doubt the value of coaching but because they cannot evidence it, and the ROI research most of the industry cites is a decade old and one-dimensional. A coach who arrives with a baseline instrument and a re-measure plan is proposing something structurally different from one who arrives with testimonials. innerly.me works as a white-label diagnostic layer for coaching platforms and independent coaches — talk to us about the partnership model.
What if the re-measure shows no improvement?
Then you have learned something valuable, and you should report it. A measurement system that only ever returns good news is not a measurement system, and experienced buyers read it as marketing. In practice, a flat result usually means one of three things: the coaching addressed a different competency than the one measured, the dosage was too low, or the engagement was mismatched to the person. Each of those is actionable, and none of them is visible if the only instrument is a satisfaction survey.
Stop selling trust.
Start selling evidence.
Whether you buy coaching or deliver it — if an engagement starts this quarter, the window to baseline it is open until session one.