Question it answers
A health system has fixed an access or cost problem. When does the fix become something a payer will build a product around?
A payer can sell a health system's solution once members are assigned to it, the plan's design moves members toward it, and a contract pays for what it delivers. Until then, the fix improves the system's own operations and gives the payer no reason to partner.
How to use it
Ask three questions in order:
- Attribute. Can a payer assign members to the solution by a rule both sides accept? A passing answer names the rule, such as primary care selection at enrollment or a claims-based method, and the members it covers.
- Steer. Can the plan's benefit design move members toward it? A passing answer names the lever, such as a lower copay, a preferred tier, or default assignment, and the members it reaches.
- Pay. Can the payer pay for it through a contract it already runs? A passing answer names the payment, such as a per-member fee or shared savings, and the evidence behind the price.
Each answer depends on the one before it, so the first question that fails tells the system what to build next.
In practice
A clinically integrated network turned a primary care access bottleneck into a virtual-first care model and a commercial HMO experience, co-developed with a regional payer and launching in January 2027. The virtual primary care practice gave a payer something to attribute members to, and that auto-attribution model became critical to the HMO. The HMO has not launched, so steering and payment are still to be proven.
Where it breaks
- Payer readiness varies by market.
- Some regional payers will co-develop a product now. In other markets a system will build for longer before a payer engages, and the test sets what to build in the meantime.
- Attribution is easiest in HMOs.
- Products where members choose a primary care physician give attribution a clear rule. In open-access PPOs, which enroll half of covered workers (KFF, 2026), attribution falls back on claims after the fact.
- Steering is limited in self-funded plans.
- Seventy-one percent of covered workers are in self-funded plans (KFF, 2026), where each employer decides whether to adopt a payer's benefit design. Copay steering can also fade. Eight to twelve years into one employer's tiered physician networks, tier had no association with specialists' new-patient share (Prager et al., JAMA Network Open, 2023).
- Pay needs proof.
- A payer prices value its own analysts can check. A cost-avoidance model passes this question when it passes the Analyst Rerun Test.
What it borrows
- Attribution research. Hoangmai Pham and colleagues showed how widely patients' care spreads across physicians, which limits any single retrospective assignment (New England Journal of Medicine, 2007), and Ateev Mehrotra and colleagues showed that the choice of attribution rule changes physicians' cost profiles (Annals of Internal Medicine, 2010).
- Value-based insurance design, from Michael Chernew and colleagues, which sets cost sharing to encourage care whose clinical benefit exceeds its cost (Health Affairs, 2007). Steering applies the same tool to a setting of care.
- Tiered networks, which sort physicians by cost and quality and set lower copays for the better tiers. When first introduced, they moved patients toward higher-value physicians (Prager et al., 2023).