The problem
The 2026 plan year is a shopping shock. The enhanced ACA premium tax credits expired at the end of 2025, marketplace premiums jumped, and auto-renewal quietly became the expensive default. On the Medicare side, plan exits and benefit cuts are pushing seniors to actually compare during the annual enrollment window. The result is more members needing real advice than brokers and navigators can hand-work.
And the tools do not help much. Plan finders rank by premium and stop there. Broker spreadsheets go stale the day a formulary changes. Carrier tools are conflicted by design. The differences that actually cost people money (a formulary exclusion, a network gap, a prior-auth trap) are buried in benefit documents nobody reads.
How it works
You hand the advisor a member (current coverage, medications, doctors, expected utilization, budget) and a candidate plan set. From there the span does the work:
- verifies the member's current coverage as the baseline (an eligibility check, not self-report), behind a consent gate that fails closed,
- retrieves the candidate plans' own policy and benefit documents,
- compares the documents side by side on the axes that matter (premium, deductible, max out-of-pocket, drug formulary, provider network, prior-auth burden),
- analyzes what the member gains or loses per candidate against their declared needs,
- ranks the candidates, with the gaps that drove each rank attached,
- renders a plain-language advisory report with decision caveats, PHI de-identified on the way out.
The whole thing is composition on AYA's verification spine (six existing capabilities wired together as data, four of them the same rails already carrying our insurance-discovery product).
What you get
A plan advisory report per member: the verified baseline, the candidates ranked against that member's medications, providers, utilization, and budget, the per-candidate gains and losses with the plan-document evidence behind each, and explicit caveats where a document was unavailable. Advice you can defend line by line, to the member and to a regulator.
Who it is for
Independent Medicare brokers and agencies, ACA navigators and enrollment assisters, employer-benefits and retiree-transition advisors, and digital-health enrollment platforms. It lands hardest with brokers who live under CMS marketing scrutiny and want recommendations with receipts, and it sits right next to the Luigi eligibility and coverage work already in the healthcare line.
Pricing
Pricing is a set of hypotheses we are validating, not a committed price sheet.
- Per-member advisory, one report per enrollment decision, for brokers, navigators, or a consumer one-off.
- Book of business, a seasonal batch across your whole book, so every member gets a verified-baseline ranked report before the window opens.
- Platform, continuous advisory wired into an enrollment platform, re-verified and re-ranked on every plan-year, formulary, or subsidy change.
Proof (dogfood)
We run the advisor on ourselves first. AYA feeds its own synthetic member panel (eligibility test fixtures plus a synthetic candidate-plan document set) through the exact deliverable a customer would buy: verify the baseline, compare the documents, analyze the gaps, rank, render. All of it as a structural dry run with no real member data and no live payer call.
What is actually true today: the deliverable pattern (AYA-3SPN-LUIGI-MEDICARE-ACA-PLAN-ADVISOR-v1) is on disk, PQS-scored at 91 (Grade A), and all six of its composed capabilities resolve to existing atoms, four of which are the same substrate already carrying the sibling Insurance Discovery product.
Honest note
This is a structural build, not a live-customer result, and this is decision support, not licensed insurance advice. Enrollment and disenrollment are deliberately outside the span, and any human distributing the output remains subject to CMS marketing and state broker rules. When a payer's plan documents are unavailable, the report declares the gap instead of silently omitting it. Ranking quality on real formulary and network data is unmeasured until a design-partner book run, and real runs need consented member profiles and a booted system. We would rather tell you that than rank plans by premium and call it advice.
*This page is a specification. The capability it describes is not built yet, and nothing here is a claim that it runs today.*