How to show group coaching results to clients and sponsors

A good coaching results report answers one question: is each member measurably different at the end than at the start? The temptation is to report activity instead — attendance, posts, sessions held — because activity is easy to count. The Kirkpatrick four-level model, the evaluation framework most corporate learning teams have used since Donald Kirkpatrick first set it out in 1959, exists precisely to stop that substitution: it puts behaviour change and results above reaction and learning, because a busy programme that changes nothing has failed however lively the group felt.

What should a group coaching results report actually measure?

Three things, in this order: where each member ended up against where they started, how the cohort behaved as a group, and what changed in plain words.

The first is the spine of the report. At week one, every member should have written a single outcome they want — a goal in their own sentence, not a target you assigned. The report shows, member by member, how far that goal moved. If a member set "run 5k without stopping" and finished the programme running it, that is a result. If they set "sleep seven hours four nights a week" and went from one night to four, that is a result. The point of self-set goals is that progress against them is unarguable.

The second is the cohort signal. One or two numbers that describe whether the group did the work: weekly check-in response rate, task completion, the proportion of members still active at the end. These show the programme was alive without pretending liveliness is the outcome.

The third is a short narrative. Two or three sentences naming the pattern — what most members struggled with, where the group turned a corner, which week the momentum built. A sponsor reads the numbers in ten seconds and the narrative is what they actually remember.

Why do output metrics make a results report worse?

Because they answer a question nobody is paying you to answer, and they crowd out the one that matters.

An output is anything that counts activity: posts, messages, sessions, log-ins, attendance. They are seductive because they are always available and always look healthy — a group of fourteen people will generate hundreds of posts whether or not a single person changed. Lead a report with "2,400 messages exchanged" and you have told the sponsor the group was busy, which they will quietly translate into "and I still don't know if it worked".

Here is the opinion, and we will defend it: leading with engagement metrics is not just weak, it is actively harmful, because what you report becomes what everyone optimises. Put message volume at the top of the page and your next cohort will chase chatter — more prompts, more reactions, more noise — at the expense of the quiet member who did the work and posted twice. A coach who reports completion against goals gets a group that completes goals. A coach who reports activity gets activity.

This is the same logic behind measuring the behaviour and not only the outcome, which we set out in our guide to progress tracking. Outputs are not worthless — a collapsing check-in response rate is a real early warning — but they are diagnostics for you, not headlines for the sponsor. Keep them in an appendix.

How do you report individual progress without breaking confidentiality?

The rule is simple: named progress goes only to the person it belongs to, and the cohort goes to everyone else in aggregate.

A member sees their own goal card and their own movement. A paying sponsor — an employer, a club, a head of department — sees the group average and the spread, never a per-name league table of who improved most and who lagged. The moment members suspect a named ranking is going to their manager, honest disclosure stops, and an honest check-in is the thing the whole programme runs on.

There is one legitimate exception, and it has to be handled at the front. Sometimes a line manager genuinely needs to see an individual's progress — in a formal development programme tied to a review, for instance. That is fine, but only if every member is told at joining, in writing, exactly who will see what. Decide it in the joining agreement before anyone discloses anything, never after. The default stays aggregate-out, named-in; the exception is opt-in and visible.

In Bitir this maps onto the product directly. Each member's goal card and progress are visible to that member and the manager, while the cohort view a manager exports for a sponsor is the aggregate — response rates, completion, active count — with no individual named unless the group's settings and joining agreement explicitly allow it.

What does a one-page results report look like?

Take Tom Beresford, who runs a 10-week resilience programme for 14 first-line NHS team leaders in Sheffield, commissioned and paid for by the trust's organisational development team. The OD lead is his sponsor, and she has three other programmes competing for next year's budget. His report has to earn the renewal in the time it takes her to read one side of A4.

His report has four blocks:

1. One headline outcome. "11 of 14 leaders met or exceeded the personal resilience goal they set in week one." One sentence, top of the page, in bold.
2. The cohort signal. Three numbers: 87% average weekly check-in response, 84% of weekly actions completed, 13 of 14 still active at week 10. Enough to show the programme held together.
3. The aggregate shift. A single before-and-after on the one measure the programme targeted — here, an average of self-rated weekly stress, moving from 6.8 to 4.2 out of ten across the cohort, shown as a small chart, no names.
4. The narrative. Three sentences: the group struggled most in weeks three and four when workload peaked, momentum built once members started logging a single daily action rather than aiming for a perfect week, and the three who fell short all cited a major service incident that month.

That is the whole report. No message counts, no screenshots of a lively feed, no fourteen-tab spreadsheet. The OD lead can see in one read that the programme changed behaviour, held its people, and moved the one number she cares about. The detail — per-member breakdowns, weekly logs — sits behind it for anyone who asks, but it does not lead.

How often should you report, and to whom?

Twice for a fixed-length programme: once at the midpoint, once at the end.

The midpoint report is the one most coaches skip, and it is the one that earns its keep. A report at week five of a ten-week programme catches a cohort that is drifting while there is still time to change course — the same week-three to week-five danger zone where most drop-out happens. Send it to yourself first as a diagnostic, then a lighter version to the sponsor if one is paying.

The end report closes the work and opens the next conversation. It is where the renewal lives — a sponsor who has just read that 11 of 14 leaders hit their goals is in a very different frame than one you call cold in March asking for next year's budget.

For open-ended or ongoing corporate programmes, a short monthly one-pager strikes the same balance: frequent enough to stay credible, rare enough that nobody starts reacting to weekly noise. Reporting every week to a sponsor sounds diligent and quietly trains everyone to treat normal variation as a crisis.

The work of building these reports is far smaller when the underlying data already lives in one place. A coach exporting goal progress, completion, and check-in response from a single cohort spends minutes on a report that takes hours to assemble from a WhatsApp group, a spreadsheet, and memory. The executive coach in our David Holt case study rebuilt his entire sponsor reporting around exactly this shift, replacing quarterly 360-degree reports with a running record he could summarise on demand.

Related reading: progress tracking for coaching groups: what to measure and how, and how an executive coach kept 11 of 14 managers through a 10-week programme.