Top-Down vs. Consumer-Up

Every market analysis in senior living starts somewhere. Where it starts decides what it can tell you.

Most of the tools the industry relies on start at the top. They take a geography, a metro or a county or a radius, and model the population inside it: how many people are over 75, what they earn, what their homes are worth. Then they estimate their way down toward a single community.

Those tools are good at what they're built for. If you need a broad read on a region's demographics, a top-down model is a reasonable place to look.

But there's a set of questions a top-down model can't answer, because it never looked at the people who actually made the decision.

What top-down can't see

Ask a modeled dataset who moved into the communities in your market last year. It can tell you who might have. It can tell you how many people like them live nearby. It can't tell you who did.

That gap matters more than it sounds. It means it can't answer questions like these:

• Where did the residents in your market actually move from?
• How far did their families live from the community they chose?
• Which nearby communities are actually winning the same households you want?
• What could the residents at each competing community really afford?

Without those answers, teams fill in the blanks with what they have. A radius around the building. The ZIP codes marketing has always used. A comp set of the five closest buildings. None of that is unreasonable. It's just never been tested.

Starting from the consumer

Occulytics works in the other direction. We start with the individual resident and their household, observed in actual consumer data, and build everything else on top of that.

That data comes from more than 100 independent sources, including mobile device data, shipping records, voter registration and utility data. Together it gives us visibility into more than 22,500 senior living communities, around 2 million residents, 8 million move-ins and 65 million records on the adult children, spouses and close family who influence the decision.

Because the picture starts at the person, it answers the questions top-down data can't. You can see where a community's residents actually came from, not where a radius assumes they did. You can see what they could afford. You can see which family members were nearby when they made the move.

Nobody grades their own homework

How the data is sourced matters as much as which direction it's built.

No operator, investor or client contributes to the Occulytics dataset. That has two consequences.

First, it's unbiased. No participation agreement shapes what's in it, and no one is reporting on their own performance.

Second, it's symmetrical. We see a competitor's residents as clearly as we see yours. That's what makes it possible to measure competitive overlap and move-in market share at all. A dataset built from operator surveys only knows what operators chose to share.

What this changes in practice

Consumer-up data doesn't replace judgment. It gives judgment something solid to stand on.

For an operator, that might mean redrawing a trade area around where move-ins actually come from, or dropping a "competitor" that never competed for your residents in the first place. For an investor, it might mean testing a pro forma's demand assumptions against real move-in behavior before committing capital.

In both cases, the decision starts from what happened, not from what a model estimates should have happened.

A simple test for your own data

Pull up the market report behind your last big decision and ask one question of it: does this show me who actually moved in, or who probably lives nearby? Then check it against your own records.

Take your last 25 move-ins. For each one:

1. Find their previous home address in your CRM or admission paperwork.
2. Put it on a map.
3. Note where their adult child or main family contact lives.
4. Lay your report's radius or ZIP list over the top.

You'll likely find residents who came from outside the lines your report drew, and family locations that explain moves the radius never would.

You'll also find the limits of doing it by hand. Twenty-five records takes real time, and it only covers your own building. Your competitors' residents are invisible from where you sit, and so are the families who considered you and chose someone else. Market Intel shows the whole market from the consumer up, for every community in it. Onboarding takes minutes, and most teams see their first answers the same day.

‍