Article
Executive Summary
- Employers now run a dozen or more point-solution vendors on average — each one solving a real problem for a slice of the population. The gap isn't the vendors; it's that none of their data connects, so nobody can see the full picture of what's working.
- 44% of employers say vendor reporting doesn't give them employer-specific ROI, and a third still can't get complete claims data from their own plan's TPA — despite a 2021 federal law intended to guarantee that access.
- Claims data is retrospective by construction — commercial payers typically allow 90 to 180 days just to file a claim, before adjudication and reporting even begin. Encounter-level data generated at the point of care is available immediately, months before the same event would surface in a claims report.
- Data has to do two jobs: measure whether the tools already in place are delivering their intended effect, and reveal, at the population level, what's still missing or needed. Most benefit ecosystems can currently do neither well, because each system's data lives in its own silo.
- The financial impact of good data shows up twice — in knowing where to direct resources, and in being able to prove, afterward, whether that spend actually contained cost. Most employers can only ever see one half of that loop.
The Data Should Already Be There — It's Just Siloed
Most self-funded employers aren't short on data, or short on good vendors. Between the medical plan, the TPA, the pharmacy benefit manager, and however many point solutions have been layered on for chronic disease, MSK, mental health, and wellness, there's no shortage of dashboards — and each of those tools was very likely selected because it solves a real problem. The average large employer now runs well into double digits worth of point-solution vendors, and industry surveys put HR leaders spending roughly a third of their average workweek just managing those vendor relationships. The problem isn't the vendors or the data itself. It's that each system holds its own slice, in its own format, with no connective layer between them — so no single view exists of what's actually working, for the employer or for the vendors themselves.
There's a second, separate limitation worth naming: even where that data is fully integrated, it's still old by the time it arrives. A claim isn't filed the day care happens — payers typically give providers 90 to 180 days to submit it — and adjudication and reporting cycles add further lag before it reaches an employer's dashboard. That's not a flaw in any one TPA or vendor's process; it's inherent to how a claim is generated in the first place. Encounter-level data, by contrast, exists the moment care is delivered. Integrating siloed data solves the visibility problem described above. It doesn't solve the timing problem — for that, the data has to originate somewhere other than a claim.
Data Has Two Jobs — and Most Stacks Can Only Do One
The first job is measurement: is the tool doing what it was bought to do? This is where the fragmentation shows up most clearly — not as a vendor shortfall, but as a visibility problem. Only about 30% of digital health solutions can demonstrate measurable ROI when evaluated on their own, and 44% of employers say their vendor's own reporting doesn't give them an ROI figure specific to their population. That's often less about the program underperforming and more about no one — including the vendor — having the full claims picture needed to prove it worked.
The second job is population-level insight: given everything the plan already knows about its people, what's still missing? A benefits strategy built from disconnected vendor dashboards can't answer that question, because no single dataset has the full picture. Knowing where a population's risk concentrates — and whether the current lineup of programs is actually addressing it — requires the data to be integrated in the first place, not just collected.
Why “Transparent” Should Be the Default, Not a Differentiator
Employers are also the ones who ultimately pay for this fragmentation, which makes it worth asking why full data access is still something to be negotiated rather than assumed. Many TPA and vendor contracts historically included clauses barring the employer's own claims data from being shared back to them. Federal law banned that practice in 2021 — yet a third of employers still report incomplete claims data access, and four in ten say a vendor has refused outright. The gap isn't academic: employers with complete data access deploy nearly four more cost-containment strategies, on average, than employers without it. Full visibility into a self-funded plan's own data shouldn't be a premium feature or a point of negotiation with each vendor. It's the baseline a fiduciary needs just to do the job — and it's a standard that benefits every party at the table, including the vendors whose good work currently has no way to get proven.
The Financial Case for Integration, Not Just Access
This is where the two functions of data meet their financial payoff. Targeted case management — support directed at the specific people who need it, informed by real population data — has been shown to reduce hospital admissions and ER visits by up to 30–40% and lower overall costs by up to 15%. But that only works if the data identifying who needs it is connected to the data confirming whether the intervention actually landed. Resource allocation without a measurement loop is a guess with a budget attached. Measurement without integrated population data is a report nobody can act on. The financial value only shows up when both halves are connected — the same dataset informing where to spend, and later confirming whether that spend worked.
An Illustrative Comparison: Two Employers, Same Vendor Lineup
Consider two employers, each running a similar mix of ten-plus point solutions alongside their core medical plan. The first manages each vendor relationship separately — a login and a quarterly report for each — with no shared view connecting any of it back to claims. When renewal comes, the MSK program and the diabetes management program both report strong “engagement,” and both get renewed by default, because there's no way to tell which one, if either, actually reduced a claim.
The second employer requires every vendor's encounter and outcome data to flow into one integrated view alongside its medical claims. At renewal, the data shows the MSK program correlates with a real reduction in surgical claims among enrolled members — worth keeping, worth expanding, and worth the vendor being able to point to as proof of their own value. The diabetes program shows high engagement but no measurable claims impact — a candidate to cut, or to work with the vendor to fix. Same number of vendors, same total spend going in. Only one employer — and one set of vendors — can actually prove which dollars are working.
How This Connects to the Rest of a Benefits Strategy
Data and transparency is the connective tissue for everything else in a benefits strategy, not a standalone reporting function. It depends on Care Delivery to exist at all — encounter-level detail only gets generated where there's a consistent clinical relationship producing it visit by visit. It's what makes Risk Sharing credible — a guarantee is only as good as the data used to audit it, for us and for any partner willing to be measured. And it's the foundation Clinical Partnership & Steerage runs on: a provider can only direct a patient, or an employer can only direct a budget, toward what the integrated data actually shows is needed.
Partner Medical's model is built around encounter-level, employer-owned data specifically so that its own impact — and the impact of everything else in an employer's benefits stack — can be measured on one shared, transparent set of numbers. In practice, that also makes us a better partner to the vendors already in place: we often see low utilization on a good program early enough to help fix it, and we support enrollment and implementation pushes for point solutions directly, because a vendor that's actually being used is good for everyone — the employer, the employee, and the vendor.
For brokers, this is a tool for the whole book, not a critique of it: integrated data gives you an evidence-backed answer at renewal for every vendor on the plan, not just ours — which is what most vendor conversations are missing today, and where a broker's recommendation carries more weight.
Contact us to learn more about how Partner Medical uses data & analytics in cost containment for self-insured stakeholders.
Partner Medical · info@partnermedical.org
References
1. Fijoya, “What are the symptoms of Point Solution Fatigue?” citing WTW survey data (44% lack employer-specific ROI reporting; 48% lack internal resources to manage point solutions).
2. e-CareManagement, “The magnitude of point solution fatigue in healthcare,” citing Quantum Health survey data (HR leaders spend ~1/3 of the workweek managing point-solution vendors; 69%/60% employer concern on siloed care and integration).
3. BSI Corporate Benefits, “Tackling Point Solution Fatigue: A Smarter Way to Manage Digital Health,” citing industry data on measurable ROI for digital health solutions implemented in isolation (~30%).
4. “The Missing Piece: Why Employers Still Can't Solve The Health Care Puzzle.” Health Affairs Forefront, citing National Alliance of Healthcare Purchaser Coalitions, Pulse of the Purchaser survey.
5. “Employers With Claims Data Access Take More Cost Action, Survey Finds.” AJMC, citing National Alliance of Healthcare Purchaser Coalitions survey.
6. “Using Computational Approaches to Improve Risk-Stratified Patient Management: Rationale and Methods.” NCBI/PMC, citing case management outcome ranges (30–40% reduction in admissions/ED visits; up to 15% cost reduction).
