Nobody buys analysis. They buy it the week a decision lands.
So this page isn’t a product list. It’s the six moments where employers call us, written the way they describe them. If one of them is on your desk right now, that’s the conversation to have.
Six times the arithmetic starts to matter.
None of these require anything to be going wrong. They’re ordinary points in a benefits year — the difference is that at each one, a number is about to be accepted without anyone independent having checked it.
Renewal is 60–90 days out
CFO · Benefits leader
“The recommendation arrived as a slide. Where’s the arithmetic?”
A renewal packet arrives with a proposed structure, a rate, and a narrative. What it rarely arrives with is the model underneath — what was assumed about engagement, which savings are already being counted by another vendor, and how sensitive the whole thing is to the two or three numbers nobody wants to name.
An independent read on the recommendation before you sign it, with the assumptions listed and adjustable.
A point solution is under evaluation
Benefits leader · CFO
“Their study says 3:1. Does that transfer to us?”
MSK, GLP-1, fertility, diabetes, behavioural health — the categories with the loudest evidence are the ones where selection bias and double-counted value do the most work. The study is usually real. The question is whether the population that produced it looks anything like yours, and who else you’re already paying to touch the same member.
The vendor’s claim adjusted for selection, overlap and evidence transfer — as a range, with their unadjusted figure shown alongside.
You’re running an RFP
Benefits leader · Procurement
“Four vendors, four sets of savings. They can’t all be right.”
Responses land in different formats, quoting different baselines, over different time horizons, each claiming the full value of an outcome the others also claim. Scored by committee, the vendor with the best narrative wins. Nobody de-duplicates the claims, and the arithmetic across the shortlist frequently exceeds the spend available to save.
A normalised comparison across responses, with overlapping claims separated and each figure’s evidence quality scored on the same scale.
You’re changing broker or consultant
CFO · Owner
“How do I judge the new recommendations against the old ones?”
A broker change resets the advice but not the baseline. Without an independent measure taken before the transition, the incoming recommendations are evaluated against nothing, and the first renewal under new advice is impossible to attribute.
A baseline portfolio assessment you own — independent of whoever is advising you this year or next.
Stop-loss renewal or attachment point
CFO · Finance
“Are we buying the right amount of protection?”
The attachment point is one of the few benefit decisions that is purely financial, and one of the least often modelled independently. It is usually inherited from last year, adjusted at the margin, and quoted by a party whose revenue moves with the answer.
The attachment decision modelled across scenarios, with the expected case stated as a range and the tail explicit.
The CFO asked what the spend is buying
Benefits leader
“I need an answer that survives a finance meeting.”
This is the moment the two conversations finally meet — and the benefits team is asked to defend a portfolio in language it was never asked to build one in. The programs may well be sound. The problem is that the case for them lives in vendor decks and lived experience rather than in a model anyone in finance recognises.
A portfolio score across eight dimensions, with the reasoning written out so it can be handed to someone who will argue with it.
End to end, on what you already have.
Most benefits analytics begins with a data project: feeds to build, carriers to coordinate, a warehouse to populate, months before anyone sees an answer. We don’t work that way. Most of what decides whether a claim holds up isn’t in your claims file — it’s in the study design, the contract terms, the overlap with what you already run, and who actually works for you.
Those live in documents on your drive. Send those, and the analysis runs. If you already have a warehouse or an analytics vendor, we work on top of it rather than against it — we’re the layer that reads the output and tells you what to do, not another place to put your data.
Nothing to install, no feed to build, nothing for IT to schedule. If you never talk to us again, nothing breaks.
Vendor decks, renewal packets, benefit summaries, a workforce profile. Aggregate and de-identified — we don’t ask for member-level data, and the intake is built so you can’t accidentally send it.
Broker, consultant, analytics platform, care management — none of it needs to be displaced for this to be useful. Independence is the product; replacement isn’t.
The same report reads three ways.
A benefit decision is usually made by people with different incentives, different vocabularies and different definitions of a good outcome. One of the quieter things an independent analysis does is give all three the same set of numbers to argue over.
CFO or owner
Benefits are frequently the second or third largest line on the P&L and the only one with no independent analytical layer. Every other spend of that size gets a second opinion as a matter of course. This one gets a renewal meeting.
You’ve started asking benefits questions in finance language and getting answers in benefits language.
Benefits or HR leader
Not a threat to your judgment — a source for it. The leaders who bring us in are the ones already sceptical of the decks they’re handed and looking for numbers that hold up when someone senior pushes back. We work alongside brokers routinely.
You believe the recommendation is right and you want something more than belief to say so with.
Broker, consultant or health plan
Independence is more useful to a good advisor than it is threatening. An analysis your client can interrogate makes your recommendation more defensible, not less — particularly where the honest answer is to spend less. Bulk and white-label arrangements are available.
Your client is asking for evidence you’d rather not produce about your own recommendation.
Where this works, and where it doesn’t.
Telling you plainly when we’re the wrong call is cheaper for both of us than discovering it three weeks in.
- Self-funded or level-funded, roughly 100 covered lives and up
- A decision actually in front of you — a renewal, an evaluation, an RFP
- Someone willing to read a model and argue with it
- Fully insured with no discretion over plan or vendor selection
- Looking for a claims data warehouse or an ongoing analytics platform
- Needing an actuarial opinion, a legal clearance or a compliance certification
“Not us” is a real answer and we give it early. A report nobody can act on is worse than no report.
Point it at the decision you already have.
Free, reviewed by a person, back within 24 hours. No call attached and no obligation afterwards — if it isn’t useful you’ve lost twenty minutes and gained a benchmark.