Last updated: Aug 2026 · Reviewed by Aaraddhya Bhatalkar, Co-CEO, New Resilience.

The short version: You cannot allocate a budget you cannot measure, and most treatment centers cannot see what any channel costs them per admission. So the first move is measurement, not reallocation. Once you can see it, the decision gets simpler: fund the channels you own and that convert, which are usually alumni and referral relationships, use the channels you rent and that are fast, like paid search, to plug short-term census gaps, and stop overspending on the expensive middle, which is usually broad conferences and unmeasured business development. Attribution will never be perfect, and that is fine. Directionally right is enough to move a budget.

Start by measuring, because most centers cannot

Most centers allocate by habit. They spend on business development because they always have, they attend the same conferences because everyone does, and they buy ads when census dips. None of it is measured against what it produces, because the data to measure it does not sit in one place. The referral source gets lost between the first call and the EHR, business development activity lives in a rep's head, and no one can say what a single admission from any channel actually cost.

So before you move any money, make cost per admission by source something you can see. Log every business development touch. Keep the referral source attached to an inquiry from the first call through to admit. Track what you spend on each channel against the admissions it produced, not the calls or the clicks. You are not chasing a perfect number here, just the difference between a channel that costs a few hundred dollars an admit and one that costs several thousand, because that difference is where the decision lives.

Attribution is the hard part, so aim for directionally right, not perfect

The catch is that you often cannot say where an admission truly came from. A family hears about you from an alum, reads your reviews, checks a directory, asks an AI tool, and calls after their therapist brings you up. That is five touches over several weeks, and the family may only remember the last one. Last-click attribution, which is what a dropdown at intake produces, credits that final touch and quietly under-counts the relationship channels that did the real work, because a word-of-mouth referral leaves no click to track.

You are not going to solve this perfectly, and chasing a perfect number wastes time. What works is getting it consistent and directionally right, and correcting for the bias you already know is there. A few things make it better.

Ask the right question at the right time. Use a neutral, open "how did you hear about us," not a forced dropdown, and ask it a few weeks after admission once the crisis has passed. At intake, people report the last thing that happened. Given a little time, they describe the whole path, which often began with an alum or a referrer.

Let more than one answer count. Capture what first made them consider you and what made them call now. Even a crude two-touch view beats last-click, because it credits both the relationship that started the process and the trigger that finished it.

Keep the referral source in the system from the first call to admit. The source tagged when the inquiry came in is a record that does not depend on anyone's memory, and you can cross-check it against what the family reports and which rep touched the account.

Read it at the cohort level. You do not need to perfectly attribute every admission. You need to know, across a quarter, roughly what share came through relationships, online, and paid. Aggregate patterns are steadier than any single case, and steady is enough to move a budget.

When in doubt, give the credit to the relationship. Last-click already under-counts alumni and referrals, so a program that splits the difference will keep underfunding the channels that convert best. Bias the other way.

A way to decide: what you own versus what you rent

Once you can see the numbers, sort every channel by two things. The first is what it costs per admission. The second is whether you own it or rent it. A channel you own keeps producing after you build it and gets cheaper over time: your alumni, your referral relationships, your reputation, and the content that makes you findable. A channel you rent stops the day you stop paying and costs more every year as more centers bid for the same clicks and directory slots.

That gives you an order of operations. Fund the channels you own that also convert first, because they are both the cheapest per admission and the most durable. Use the channels you rent that are fast, mainly paid search, to fill beds in the short term when census dips, knowing you are renting and it stops the day you stop paying. And look hardest at the expensive middle, the broad conference calendar and the unmeasured business development, because that is usually where money leaves without producing admissions anyone can trace.

This is also the answer to the census question owners actually worry about. Demand you rent gives you this month. Demand you own gives you next year. A budget weighted toward what you own is a bet on a predictable census, which is worth more than a cheap admit today.

So where does that leave BD reps, referral sources, and conferences?

Referral sources are the channel to protect. Professional and alumni referrals are the lowest cost per admission a center has, and they convert several times better than cold online, because they arrive with trust attached. The mistake is treating them as a given rather than a channel. Score your sources, see which ones produce admissions rather than phone calls, and put your effort into deepening the handful that do.

Business development is how you work that channel, so hold it to the same standard. Log each rep's activity, tie it to the admissions it produced, and you will usually find that a few reps and a few sources carry most of the result. Redeploy from there. This is as much a reallocation of where reps spend their time as it is of dollars.

Conferences have changed. A decade ago, business development ran on presence: send reps to as many events as the budget allowed and let relationships build over time. The programs we talk to say that no longer pays the way it did. Everyone is at the same conferences, every referrer has met a dozen reps, and a booth and a handshake do not set you apart anymore.

What is working now is fewer conferences, chosen by intent. Rather than going broad, you go to the specific events where the referral sources who send your kind of client actually gather, the ones matched to your levels of care, your specialties, and the populations you treat. And you go as a specialist. Referrers send their clients to the program that clearly owns one thing and can prove it, so the job at these events is to come across as the expert in exactly what your program does, and to talk about your clinical outcomes with the people who understand them and refer based on them. A handful of the right conferences, where you show what your program does and what happens to your patients, will do more than a dozen where you are one more booth.

Reallocation is a people problem, so move at the margin

None of this happens in a spreadsheet. You cannot drop a tenured business development rep or pull out of the conference the whole field attends without friction, and treating it as a pure numbers exercise is how good plans stall. So move carefully. Measure for a quarter before you change anything, so the case is yours and not a guess. Then reallocate at the margin, shifting the next dollar and the next week of a rep's time toward what the data favors rather than making a dramatic cut. Point reps at the intent-based conferences and the referral relationships that convert instead of cutting heads. The goal is to steer the budget, not to blow it up.

Where to start

If you do nothing else, do these three. Make cost per admission by source something you can see, even roughly. Protect and grow the two channels that convert best and cost the least, your alumni and your referral relationships. And pick your conferences by intent, going deep where the right referrers are rather than broad where everyone is.

New Resilience gives you the part that is hardest to do by hand. It logs business development activity, keeps the referral source attached from the first call to admit, scales alumni outreach over text, and ties every admission back to its source, so you can finally see cost per admission by channel. Book a demo to see how it works.