What drives your admits?

Ask an owner what drives their business and you will usually get a confident answer. “We get a lot of calls off the website. Our BD team produces. Between those two, that’s most of it.”

That answer isn’t wrong. It just doesn’t give you anything to pull on. Nothing in it says where the next dollar, the next hire, or the next hour should go.

The second question

Ask how people find the website and the answer needs another layer. “Google” might mean an ad, a service page or a search for a name a clinician recommended. The website is a destination. The path to it matters.

A Google Ads campaign, a Business Profile, an organic service page and a directory listing are distinct touchpoints. BD can create demand that later appears as a direct call or branded search. I want those paths recorded without forcing them into a false choice between referral development and the web.

An admissions team can ask “how did you hear about us?” and get “the internet.” That answer has value, but it does not distinguish a blog post from a sponsored ad or explain who suggested the search. I preserve the caller’s account alongside the recorded path.

The phone number problem

In behavioral health it gets worse, because the person who admits is usually not the only person who called.

A mother calls first, from her cell. The father calls the next day from work. A sibling calls, then finally the client. The opportunity gets created in the CRM under one of those numbers. Every other number, including whichever one was actually the first touch, is invisible to it.

If there is no lookback window, and no mechanism to capture call transcripts or attach multiple numbers at the opportunity level, the lookback has nothing to look back to. Best case, the call ties back to the one web session that was shown that tracking number, and that assumes the caller is the person who sat in the session. Numbers get passed around. A sister finds the facility and texts the number to her mother, and when the mother calls three days later, the call is credited to the sister’s visit at best, to a stranger’s session at worst, or to nothing at all. Wrong data survives into the reports in a way missing data never does.

Across the attribution chains I’ve audited, a single phone number can miss calls from other people involved in an admission. Finding those calls is one step. Establishing which source brought the person to the facility is another.

What guessing costs

Double a paid-search budget and calls may stay flat. Add a BD territory and admissions may not rise. Before replacing a vendor or changing the comp plan, I want to know which link failed: demand, call handling, clinical fit, capacity or the measurement itself.

A new dashboard cannot repair a broken join. It can make the wrong answer easier to read.

What the wiring reveals

When a rep leaves and admissions remain, that is a reason to examine relationship continuity. Branded search shows how a family reached the facility; it does not establish who created the demand. I compare referral history, call records and the admission timeline before drawing conclusions about channel contribution or individual dependence.

The comp plan and the call report can both describe a piece of the same admission. A referral may have prompted the search. The useful question is which relationships persist through staff changes and what evidence supports changing coverage or compensation. A tracked touchpoint alone cannot answer it.

The moment it matters

Maybe none of this feels urgent. Census is strong, the program is differentiated, demand in your market is holding, referral relationships are warm. Today, you may genuinely not need it.

Then a new facility opens nearby, or demand softens. The decision gets harder when the baseline is missing: is the problem fewer inquiries, slower callbacks, a change in clinical fit or capacity? Knowing those measures before the disruption gives the team somewhere concrete to start.

What a lever looks like

A real attribution model is more than reporting. It assigns time, money, and staff effort to specific problems, with specific measurables on the other end, and gates that tell you whether each investment is working. It means the chain is wired from first touch to admit and validated against your CRM and your call center rather than a platform dashboard. It has to be built by people who know the wiring, and then maintained, because it decays the moment nobody owns it.

This work crosses boundaries. It can require changes to how admissions logs calls, how the CRM records opportunities and how a vendor reports performance. Someone has to own the shared definitions and the handoffs. That is a leadership responsibility as much as a technical one.

Where I start

I start by mapping the attribution chain as it exists and fixing the breaks that matter to the next decision. Essential growth work continues while the measurement improves; the size and reversibility of a spend decision should match the evidence available. The admission-level attribution loop on this site joins CRM and call records through stamped IDs, backlinks and time-bounded phone matches. It distinguishes call linkage from named-source coverage and flags shared numbers for review.

A repaired chain can change the old story. It should also make its own limits visible: which records linked, which sources remain unknown and which conclusions require more than attribution. I use that evidence to choose the next operating change, then look again at what happened.

That’s a lever. That’s knowing what drives your admits.