The SNF advocate idea is clever. The unit economics might kill it anyway.
There's a real problem buried inside the SNF Advocate Marketplace idea, and I think it's worth taking seriously before you dismiss it or build it.
Somebody's parent is in a skilled nursing facility. Medicare sends a notice saying coverage stops Friday. A discharge planner hands the family a pamphlet and wishes them luck. The family has 48 hours to file an appeal and absolutely no idea what that means or who can help. They Google something like "Medicare stopped paying for SNF" in a panic, and the results are a graveyard of government PDFs and AARP articles from 2019.
That's a real, recurring, painful moment. And nobody has built a decent product for it.
The insight driving this idea is actually good: high-intent search traffic, a defined regulatory window, and an invisible supply of credentialed advocates who have no way to market themselves. There are roughly 1,500 BCPA-certified advocates in the US. Most of them are working out of personal websites or word-of-mouth referrals. Families in crisis can't find them. Advocates can't find families at scale.
That gap is real. The platform that shows up in that search moment with three vetted advocate profiles and transparent pricing probably does convert. The question is whether you can build a business around it.
Here's the thing I keep getting stuck on. Medicare SNF appeals run on a statutory timeline. The window is about two weeks. After that, the moment is over. The family either won or lost, and either way, they're never buying this again.
This isn't like a legal marketplace where clients come back for estate planning, contract disputes, tax issues. SNF discharge crises are once-in-a-lifetime events for most families. The LTV on the family side is $225, which is fine, but you're paying acquisition cost on every single case with zero repeat purchase. If you're relying on organic search and Reddit posts to drive family traffic, that might pencil out. If you're running paid ads, the math gets painful fast.
The facility referral side is where the business model tries to solve this. A discharge planner at an SNF is generating these moments constantly. If you can get a referral QR code into 100 facilities at $99/month, that's recurring revenue that doesn't depend on finding a new family every time. The LTV on a facility seat ($1,782 over 18 months) is what actually makes this work as a business.
But then you run into the conflict of interest problem, which I think the idea undersells.
A for-profit SNF has financial incentives to discharge patients. That's just the business model of post-acute care. A discharge planner at one of these facilities is being asked to hand families a tool that might help them fight the facility's decision. That's not a comfortable thing for most people to do, regardless of whether it's technically allowed.
The pitch to discharge planners is framed as "protects you from being in the middle of appeals," which is smart. But I'm not sure it's enough. The social worker isn't the decision-maker on vendor adoption at a chain facility. IT security reviews, legal sign-off, centralized procurement processes, these things exist and they take 6 to 18 months per chain. If your go-to-market assumes you can email a director of social work and have a referral tool live next week, you're in for a rough education on how healthcare procurement actually works.
There's also the Medicare Advantage problem. MA plans now cover more than half of Medicare beneficiaries and have their own internal appeals processes. A product built around traditional Medicare Part A appeals is addressing a market that's structurally shrinking. Not collapsing, but shrinking. You'd want to understand that trajectory before building deep assumptions around it.
Marketplace businesses need density to work. If a family in Omaha needs an advocate within 2 hours and the platform has 3 advocates in Nebraska, one of whom is retired and one of whom is backed up with cases, the platform fails at the moment that matters most.
With only 1,500 BCPA-certified advocates nationally, many part-time, many geographically concentrated, building a marketplace that can consistently deliver 3 matched options in any metro area is genuinely hard. You're not Uber where supply can surge to meet demand. You can't spin up new advocates in a week.
The validation test the idea proposes is actually the right move here. Pull 30 advocates from the BCPA directory, run the fake door landing page, see if any cases actually close. That tells you whether the supply side will engage before you spend 7 weeks building.
If I were taking a serious run at this, I'd ignore the facility channel for the first six months. It's the right long-term business, but it's the wrong starting point. Enterprise healthcare sales with a product that hasn't proven itself is a slow death.
Instead: SEO content targeting the specific search queries families type during a discharge crisis, paired with a manually-curated advocate directory that routes intake forms to advocates who pre-commit to 24-hour response times. No marketplace software yet. Just a Google Sheet, a Typeform, and email. If advocates close cases and pay a referral fee, you have a business to build on. If they don't, you've learned that in 30 days instead of 7 weeks.
The AI angle is genuinely compelling. First-draft appeal letters generated from a 10-field intake form, calibrated to the specific denial type and payer, cut advocate prep time enough that you can justify the platform fee purely on time savings. That's where the product becomes sticky for advocates, not just a lead-gen tool they tolerate.
This is marked as "worth exploring" with a "vulnerable" survival verdict and I think that's about right. The problem is real. The market is underserved. The business model has enough pieces that it could work.
But the repeat-purchase problem on the family side, the conflict of interest on the facility side, and the thin supply of credentialed advocates are all structural constraints that a marketplace format struggles with. You're not building Airbnb. You're building something more like a specialized legal referral service for a highly specific regulatory niche that may be smaller in five years than it is today.
That's still potentially a $10M ARR business for a solo founder who gets deep enough into the domain to build real moats. But it requires being honest about what you're actually building, and what you're not.