The decisions are big.The tools to evaluate them shouldn’t be a black box.
When a benefits vendor tells you their program will save you money, Axionia checks whether that’s true — independently, with every assumption on the table. Then we tell you what the same budget could buy instead. Built for the HR leaders and CFOs who want defensible numbers and a better mix, not another pitch deck.
Free · reviewed by a person · in your inbox within 24 hours
Companies buy health programs for their employees — diabetes management, mental health support, physical therapy. The company selling the program also supplies the study proving it works. That study gets read by a lot of people and checked by almost none of them.
We check it. We take the vendor’s claim apart, adjust it for your actual covered population, and tell you what the program is realistically worth — showing every assumption we used, so you can argue with any of them.
Then we do the part that’s worth more: we tell you what else the same money could buy — including options nobody sells you, because they carry no commission and appear in no catalogue.
Anyone carrying the risk and the decision.
Headcount isn’t the qualifier. What matters is whether you’re self-funded or level-funded — so the savings accrue to you rather than to a carrier — and whether point solutions have been arriving one at a time for long enough that nobody has added them up. That describes an eight-hundred-person manufacturer and a forty-thousand-person health system equally well.
At larger scale the analysis doesn’t change; the deployment does. Organisations with data-residency, security or procurement requirements run the same agents on their own infrastructure.
You're being pitched constantly and asked to justify last year's decisions.
An independent read means you walk into the renewal with numbers your CFO will accept and a defensible reason for every recommendation — including the ones where the answer is to keep what you have.
It's one of your largest recurring costs and the one you can least interrogate.
Benefit decisions arrive as a recommendation and a vendor's ROI study. We translate them into economics you can actually challenge — ranges, assumptions, and what the claim is worth after adjustment.
You've been told the number goes up every year and that's just how it is.
Sometimes it is. Often a meaningful share of the increase is programs that overlap, contracts that trailed the market, or a claim nobody checked. We'll tell you which, without a stake in the answer.
Fifteen people reviewed it.
None of them checked it.
A benefit decision passes through more hands than almost anything else a company buys — brokers, carriers, consultants, internal committees, finance. It isn’t that nobody is paying attention. It’s that attention and scrutiny aren’t the same thing, and almost nobody in that chain is positioned to supply the second one.
Five kinds of organisation look at benefit decisions. Each is good at something, and most employers need more than one of them. None is built for this particular job.
| Who looks | Who pays them | What they're good at | Why it isn't this |
|---|---|---|---|
| Brokers and consultants | Commission and overrides from the market | Market access, placement, plan design, service. Most employers should keep theirs. | Across categories there is no common denominator and no independent benchmark to anchor to — a gap in the market rather than in their work. |
| Vendor-funded validation | The vendor being validated | Real methodology critique, often better than a buyer could do alone. | The buyer isn't the client. |
| Claims analytics | You | Showing what actually happened inside your own claims. | Looks backward at what you already bought, one program at a time. |
| Actuarial firms | You | All of it, properly. The honest answer to “who else does this”. | Bespoke, six figures, one program at a time, aimed at jumbo employers. |
| Independent assessors | Philanthropy | Category verdicts nobody can buy. | Category-level, and not about your population. |
An actuarial firm will do this for a jumbo employer at six figures, one program at a time. Nobody does it for the nine-hundred-person manufacturer, and nobody untangles double-counted value across a whole portfolio at any price.
This is a job that didn’t exist, not a job someone else was doing badly.
The costs add up. The savings don’t.
Point solutions arrive one at a time — telehealth, then MSK, then behavioral health, then weight management. Each looks trivial on the day it’s approved, and each came with a study showing it pays for itself. The costs accumulate cleanly. The savings do not: an avoided surgery can only be avoided once, and when two vendors both count it, the arithmetic across your portfolio quietly exceeds what was ever there to claim.
Not one of these looked expensive on the day it was approved. Six dollars a month is a rounding error. They were signed in different years, by different people, each against a vendor study nobody had reason to doubt.
Stacked, they run $738K a year — real money, sitting alongside your medical spend, and largely unexamined since purchase.
Some of these are creating durable value. Some are paying for outcomes you’d have gotten anyway. Almost nobody knows which is which.
Ranges are illustrative and vary widely by vendor, contract structure and covered population — larger employers typically negotiate lower per-member rates and carry more programs, which moves the total in both directions. Yours will differ. The point is the arithmetic, not the specific figures.
Costs only ever add. Claimed savings don’t — and nobody is double-charging you for that. The same result is simply being counted more than once, and no single program is positioned to see it happen.
10.6 PMPM — about 18% of everything claimed, or $89K–$120K a year at 820 covered lives.
Every one of these vendors is reporting its own results correctly. The double-count only exists once you own all five.
Illustrative figures for a self-funded employer of about 820. Double-counting is modelled as the union of independent claims against each pool — a deliberately conservative floor, since vendors target the same high-cost members on purpose and real duplication typically runs higher. Yours will differ. The point is the arithmetic, not the specific figures.
And value was never only claims cost. Every one of these programs also claims time back in the seat, fewer absence days, better productivity — against the same people, in the same year. Those claims are harder to check than a medical trend line, which is exactly why nobody checks them. Sorting out what is genuinely additive is the difference between a number you can take into a renewal and a number you can only repeat.
Your portfolio, scored on eight dimensions — against employers like you.
Eight dimensions, scored independently and plotted against the peer median (dashed). The two lowest axes — vendor independence and CFO engagement — are where this portfolio trails comparable employers, and they are the two that most often move together.
Illustrative composite profile — no cost, no commitment
Knowing what it’s worth is the start. Knowing what else it could buy is the point.
Independent analysis of what you already run tells you the real size of the budget. It doesn’t, on its own, tell you whether that budget is pointed at the right things. Those are two different jobs, and the second one is where the money is.
We do both, in that order, because the order matters. A mix proposed without an audit behind it is just another opinion.
Claims de-duplicated, selection adjusted, every program scored on one scale against your covered population. This is the part that establishes what the budget actually is, as opposed to what the invoices add up to.
The same scale ranks things you don’t currently run. Some of the strongest cost nothing at all.
A single mix rarely serves everyone equally. We model it by workforce group, so the result answers each group on the axis that group actually feels — inside one budget, not by growing it.
We didn’t place any of it, so we have nothing to defend by leaving it alone.
See where your benefit portfolio actually stands.
A few minutes to request, reviewed by a person, and in your inbox within 24 hours. It benchmarks your portfolio against comparable employers, costs nothing, and has no sales call attached to it.
No software to roll out · no data feed to build