Here's the uncomfortable part. Your client's finance lead has probably already seen the research that says programs like yours don't pay for themselves.
The most-cited study on the question came out of RAND, published in Health Affairs, tracking PepsiCo's wellness program across more than 67,000 eligible workers over seven years. It found the disease management side returned $3.78 for every $1 invested, with a 29% drop in hospital admissions. The lifestyle management side showed no significant effect on health care costs. Senior author Soeren Mattke warned buyers not to assume lifestyle components reduce costs at all.
Five metrics do that. Here's what each one is, and how to actually produce it.
1. Enrollment: did anyone start?
Enrollment is the first number your client looks at, because it's the one they can feel. If their people didn't sign up, nothing downstream matters.
The bar is lower than most operators think. More than 80% of employers offer an employee assistance program, yet according to SHRM, average utilization sits in the low single digits. Your client has almost certainly bought something nobody used.
Put that next to single-digit EAP utilization and you're not reporting a number, you're reporting a category difference.
Report it as a percentage of eligible people, never as a raw headcount. "1,200 enrolled" means nothing without a denominator, and your client will assume the worst.
2. Completion: did they finish?
Completion is where most coaching reports go quiet, because it's the number that exposes whether the program had any grip.
In Avidon's 12-month efficacy review across 23 client groups in seven industries, 73% of participants who attended their first session completed the entire four-session program. Everyone who showed up once mostly saw it through.
Track completion by cohort and by program, not as a single blended figure. A blended number hides your weakest offering, and your client will find it eventually.
3. Drop-off: where exactly did you lose people?
This is the metric that separates operators who look in control from ones who look surprised. Don't report a drop-out rate. Report where the drop-off happens.
A controlled study of 300 non-incentivized participants makes the point. Program completion ran 17% with no coaching, 28% with live coaching alone, and 36% with coaching plus technology, a 112% improvement over no coaching. But the interesting data is the session-by-session view:
| Sessions attended | No coaching | Coaching only | Coaching plus technology |
|---|---|---|---|
| Session 1 | 0 | 76 | 82 |
| Session 2 | 0 | 39 | 42 |
| Session 3 | 0 | 2 | 17 |
| Session 4 | 0 | 0 | 12 |
Coaching alone got two people to session three and nobody to session four. Add automated touchpoints between sessions and those become 17 and 12.
Sessions one and two look nearly identical across both groups. The entire difference shows up at three and four, which is exactly where behavior change gets decided. If you can show a client the specific session where your retention curve bends, and what you changed to bend it, you sound like an operator instead of a vendor. Most of that work happens between sessions rather than during them.
4. Loyalty: would they do it again?
Satisfaction scores are easy to dismiss as soft. Two versions of them aren't.
The first is voluntary re-enrollment. In that same six-year dataset, 52% of participants who completed the 12-week program asked to join a follow-on six-week program. Nobody asks for more of something that wasted their time.
The second is a recommendation rate with a real denominator, from the 12-month efficacy review:
That last detail is the one to copy. Beating a bar your client defined is far more persuasive than beating a bar you picked.
5. Behavior change: did anything actually change?
This is the number that answers RAND's objection, and it's the one most coaching companies can't produce.
Avidon's per-condition studies each surveyed participants over six months:
A separate review of 60,000+ annual health risk assessments, comparing before and after at least one online course, found 47% reported lower stress, 52% reduced their drinking, 33% quit smoking, and 53% lowered their BMI by more than 5%.
Two rules when you report this. Always attach the sample size and the timeframe, because an unsourced percentage reads as marketing. And call it what it is: these are program outcomes from participant surveys and health risk assessments, not clinical trial results. Say "reported outcomes," never "clinically proven." Overclaiming is how you lose a renewal you'd otherwise have won.
What this looks like when it works
An insurance services company selling full-service wellness programs to its own clients moved its coaching onto the Avidon platform in early 2021. It went from an in-house coaching staff to one full-time coach delivering 2,849 sessions, up from 1,700, at roughly 30% lower cost.
"Costs for the coaching went down by almost 30% while the number of coaching sessions delivered increased more than 60%. Avidon is easy to work with, and coaching participant satisfaction is excellent." — Danielle, Sr. Wellness Consultant, Insurance Company
That's a renewal conversation with nothing left to argue about. More sessions delivered, less money spent, satisfaction intact, and every figure traceable. Read the full case study.
The reporting is the product as much as the coaching is. If you're building this by exporting spreadsheets and reconciling them by hand, you'll produce it once for a renewal and never again. Avidon's reporting and ROI insights generate these five metrics continuously, so the renewal conversation is a screenshot rather than a project.
See how coaching companies and platforms run programs and prove outcomes on one platform.
