Does Corporate Wellness Actually Work? An Honest Look at the Evidence
A straight answer for execs and HRBPs weighing a wellness spend: what the ROI research really shows (it's mixed and contested), why a famous RCT found close to nothing, and the right reasons to buy a program anyway.

If you're the person who has to defend a budget line, "does corporate wellness work?" is the question that actually matters — and it's the one most vendors answer with a single tidy number. We sell wellness programming, so you'd expect us to wave a big return figure at you too. We won't. The honest answer is that the ROI evidence is mixed, sometimes flatly contradictory, and frequently contested — and the smartest buyers we meet already know that.
We run corporate sessions out of a real fight gym, founded by a professional fighter. So here's the version we'd want if we were the ones signing the PO: what the studies actually found, why two reputable sources can disagree by an order of magnitude, and what we think is the right reason to spend the money anyway.
Why the headline numbers don't agree
Search "wellness ROI" and you'll find returns quoted with total confidence — and they don't line up with each other. That's not a typo or bad faith. It's the literature genuinely disagreeing, because different studies measure different things, on different populations, with wildly different rigor.
Start with the optimistic end. Deloitte's UK analysis "Mental health and employers" (2024) reports an average return of £4.70 for every £1 spent on mental-health interventions (Deloitte, 2024). That's a real figure from a serious firm — but read what it is. It's a UK-only literature-review average, not a controlled trial, and it pools many intervention types together. A literature review aggregates existing studies; it doesn't run a clean experiment of its own. Useful as a directional signal, not as a promise you can hold a vendor to.
Now the sober end. When the RAND Corporation evaluated PepsiCo's wellness program — seven years of data on real employees — it found an overall return of about $1.50 per $1 spent (RAND, 2014). Positive, but a long way from the eye-catching figures. And the average hid something important, which is where this gets interesting.
The detail most summaries skip
RAND split PepsiCo's program into two halves, and the halves behaved nothing alike. The disease-management component — targeted help for employees already managing a chronic condition — returned about $3.78 per $1. The lifestyle-management component — the broad, everybody-gets-it wellness piece — returned about $0.50 per $1 (RAND, 2014). That second number is a loss. The program made money treating sick people and lost money nudging healthy ones.
That single split explains most of the confusion in this category. "Wellness" is not one thing. Bundle a high-return clinical component with a low-return lifestyle component, average them, and you can produce almost any headline you want depending on the mix. Anyone quoting you one number for "wellness" is, knowingly or not, hiding that variance.
The study that found close to nothing
Then there's the result the optimistic camp would rather you didn't read. The Illinois Workplace Wellness Study is the strongest design in this whole conversation: a randomized controlled trial — the same gold standard used to test drugs — run by Jones, Molitor, and Reif and published in The Quarterly Journal of Economics in 2019. Roughly 4,800 employees, randomly assigned to a wellness program or a control group, followed for two years.
The finding: no significant effect on health outcomes, medical spending, or productivity over that period (Jones, Molitor & Reif, QJE 2019). The people who looked healthier in the program were already healthier before it started — they were the type who sign up, not people the program changed. Earlier, rosier estimates had been fooled by exactly that selection effect, which a randomized design strips out.
We're not citing this to talk you out of buying anything. We're citing it because it's the most rigorous evidence available, and a serious case for spend has to survive it. If a vendor's pitch falls apart the moment you mention the Illinois RCT, that tells you something about the pitch.
So where does that leave the global figures?
You'll sometimes see wellness defended with returns from the World Health Organization. Worth being precise here too. The WHO-backed analysis led by Chisholm and colleagues, published in The Lancet Psychiatry in 2016, estimated returns of 2.3 to 3.0 to 1 in economic and productivity terms — rising to 3.3 to 5.7 to 1 once you add the value of better health itself (Chisholm et al., Lancet 2016).
Those are encouraging numbers, and they're real. But note what they measure: scaling up depression and anxiety treatment across entire health systems in 36 countries, 2016–2030 — not the return on a corporate wellness program. It's the wrong unit to bolt onto a procurement deck. We mention it because you'll see it misused, often rounded to a single tidy "4 to 1." The primary source gives two ranges and a specific context. Treat it as evidence that mental-health treatment pays off at the population level, which is a different claim from "this office perk pays for itself."
What this means for your decision
Put the four together and the picture is honest, if less satisfying than a single number:
- Optimistic literature reviews (Deloitte, £4.70:1) point one way — directional, UK-only, not experimental.
- A careful real-world evaluation (RAND/PepsiCo, $1.50:1 overall) says modestly positive on average, with the return concentrated in clinical disease management ($3.78:1) and the broad lifestyle piece actually losing money ($0.50:1).
- The cleanest experiment (Illinois RCT) says close to nothing over two years.
- Global treatment economics (WHO/Lancet) say mental-health care pays off — but at the level of health systems, not office programs.
The honest read isn't "wellness is a scam" and it isn't "wellness is a guaranteed return." It's that the financial case is genuinely contested, the returns depend heavily on what you actually buy, and anyone selling you certainty is selling past the evidence. We collected these figures on our corporate numbers page precisely so you can pressure-test any pitch — including ours — against the primary sources.
Then why buy a program at all?
Because financial ROI is the wrong first question for most of what HR actually buys. A few better reasons hold up regardless of where the ROI debate lands:
You have a specific, named problem. Notice the one component that reliably paid back in the RAND data: disease management — targeted help for a defined need. The lesson generalizes. Wellness spend tends to do its best work when it's aimed at something concrete, not sprayed across the org as a generic perk. If you can name the problem, you can choose the right tool and judge whether it worked.
You're buying participation, culture, or retention support — and you say so. A team session that gets people out of their chairs, moving and laughing together, can be worth running for what it is: a real shared experience, a signal that the company invests in its people, a break in a grinding quarter. Those are legitimate purchases. They just shouldn't be dressed up as a hard-dollar return they can't prove. Be honest about what you're buying and you'll buy better.
It complements your real systems — it doesn't replace them. No wellness program fixes a workload problem, a bad manager, or a broken policy. If burnout is structural, the fix is structural, and a class on the calendar won't substitute for it. A program sits on top of those systems; it isn't a stand-in for them.
That's the frame we bring to corporate work, and it's why we won't quote you a return on our own sessions. We can tell you exactly what a session is and does — a controlled, beginner-friendly team workout with no live sparring, built and run by professional coaches. You can see what that looks like on the corporate page. What we can't honestly tell you is that 90 minutes on the mat will move a number on a spreadsheet, and we'd rather say so than pretend otherwise.
How to decide like a skeptic
A few questions separate a defensible spend from a hopeful one:
Name the outcome before you buy. Participation? Retention support for a specific team? A morale lift after a hard stretch? Write it down first. If the only stated goal is "ROI," you've already drifted into the part of the evidence that's weakest.
Ask vendors what they're actually claiming. A provider who promises a productivity percentage or a turnover reduction from a single program is, on the evidence above, overreaching. A provider who tells you plainly what the session delivers — and what it can't — is one you can trust on the rest.
Match the tool to the problem. Broad lifestyle programs are the part of the literature that most often fails to pay back. Targeted, voluntary, well-run interventions aimed at a real need fare better. Our team-session calculator helps you frame the actual cost of the problem you're solving, so the spend is sized to something real rather than to a hoped-for return.
Budget it as what it is. A culture or wellbeing line item judged on participation and feedback is honest. The same spend justified by a borrowed ROI figure is a claim you may have to walk back. Pick the framing you can defend in the room.
Decide that way and the question stops being "does wellness work?" in the abstract — a question the evidence can't answer cleanly — and becomes "will this specific thing, for this specific reason, do what I'm buying it for?" That one you can answer. When you want to scope it, you can talk it through with us directly.
Frequently asked
Does corporate wellness actually have a positive ROI? The evidence is mixed and contested. Deloitte's 2024 UK literature review reports an average of £4.70 per £1, but a careful real-world evaluation of PepsiCo by RAND found about $1.50 per $1 overall — and the cleanest study, the randomized Illinois Workplace Wellness Study (2019), found close to no significant effect on health, spending, or productivity over two years. There is no single honest "the ROI is X" answer.
Why do different sources report such different returns? Because "wellness" isn't one thing and the studies differ in rigor. In the RAND/PepsiCo data, the disease-management component returned about $3.78 per $1 while the broad lifestyle component returned about $0.50 — a loss. Average those together and the headline depends entirely on the mix. Literature reviews also pool many programs, while a randomized trial isolates the effect of one.
Should we skip wellness programs, then? Not necessarily — just buy them for the right reasons. The most defensible spend targets a specific, named need (the part of the evidence that reliably paid back) or is bought openly for participation, culture, or retention support rather than a promised hard-dollar return. We won't quote a return on our own sessions; we'll tell you exactly what a session is and does.
What can a team session like yours honestly deliver? A controlled, beginner-friendly, coach-led team workout with no live sparring — a real shared experience, movement, and a break from the desk. We don't claim it raises productivity or lowers turnover, because we can't prove that and the evidence wouldn't support it for any single program.
How should we budget a program if the ROI is uncertain? Put it on a culture or wellbeing line and judge it on participation and feedback, not a borrowed return figure. Name the outcome you want before you buy, match the tool to that outcome, and you'll have a spend you can defend.
Start a corporate program at KD MMA
Founded by WEC veteran Karen Darabedyan, KD MMA runs corporate sessions on-site at your office or at our Glendale academy — controlled, beginner-friendly, and honest about what they deliver. If you want a program scoped to a real outcome instead of a borrowed ROI claim, we'll help you build one and tell you plainly what it can and can't do.
Estimate your team's number · request a quote · or call us at (747) 231-5550.
Give them a summer that counts.
Register early and save 10%. One week or both — spots are limited.