Cash Flow Forecasting for Therapy and ABA Practices: A 13-Week Forecast You Can Run Yourself

A 13-week cash flow forecast is a one-page spreadsheet that projects your bank balance week by week for the next quarter: expected collections in, expected payroll and bills out, projected balance at the end of each week. It answers the one question your bank balance can’t: will there be enough cash in the account on the weeks that matter? In a therapy or ABA practice, the weeks that matter are payroll weeks.

This piece walks through why the bank balance fails as a forward view, how a summer scheduling dip turns into a September cash problem, and how to build and run the forecast in about an hour, then ten minutes a week. There is a free template at the end so you don’t have to build the spreadsheet yourself.

What the bank balance tells you, and what it leaves out

Most practice owners I work with run the business from the bank balance. It is a reasonable instinct. The balance is accurate, it updates daily, and it requires no bookkeeping to read. The owner of one ABA agency told me she checked the account every morning before deciding anything else about her day, because whether one payer’s deposit had landed determined what the week was going to feel like.

The problem is that the balance only describes today. It says nothing about the payroll run in nine days, the rent that clears next week, or the fact that collections are about to soften because of something that happened to your schedule a month ago. Practice costs are lumpy. Payroll lands every two weeks and is usually the largest single outflow by a wide margin. A balance that looks comfortable the morning after a strong deposit can be two scheduled outflows away from a problem, and nothing on the screen distinguishes those two situations.

Each financial view answers a different question, and none of the standard ones look forward:

How a July session dip becomes a September cash problem

Summer compresses the schedule at most outpatient practices. Clients take vacations, clinicians take vacations, school routines that anchor family calendars disappear. Across the practices I work with, July and August typically run 10 to 20 percent fewer completed sessions than a normal month. For ABA agencies the summer shape depends on program mix, since school-based hours drop while center schedules shift, but the mechanics below apply the same way to whatever shape your calendar takes.

The part that catches owners is the delay. Insurance pays 15 to 45 days after the session, depending on the payer. So the cash from a soft July doesn’t go missing in July. It goes missing three to six weeks later, in late August and September, when the deposits that would have come from those sessions fail to arrive.

Run the arithmetic on a mid-sized group practice: $40,000 in a normal week of collections, and a 15 percent session dip across eight weeks of summer. That is roughly $48,000 of revenue that will quietly not show up, spread across the weeks of late August and September, right as payroll keeps landing on its usual schedule. Meanwhile the July bank balance looks fine, because July’s deposits are paying for June’s full schedule. The owner reads September as a surprise. It was visible in the schedule two months earlier; nothing they look at was built to carry the signal forward.

Why your books don’t warn you

The warning fails to arrive because of how the systems are wired. Your EHR knows sessions were delivered but nothing about your costs. Your billing platform or clearinghouse knows what has been claimed and paid but not what payroll runs next week. QuickBooks knows what happened after the bank does, and if a bookkeeper closes your books on a normal cycle, you get July’s story in the middle of August.

Nothing in that chain projects forward, and the three systems don’t talk to each other in a way that produces a forward view of cash. This is not a software gap you should feel behind for not having solved. It is the normal state of a practice this size. The fix is not another integration; it is a small spreadsheet that sits above all three and does one job.

The 13-week forecast, line by line

I build a version of the same forecast in nearly every engagement, whether the practice is a five-location counseling group or a single-site ABA agency. It is deliberately small. One page, weekly columns, and about ten rows:

1. Starting balance: the bank balance at the start of the week. Week one is typed in; every later week carries forward automatically.
2. Insurance collections: expected deposits from payers, your largest and least certain line.
3. Patient payments: copays, self-pay, patient balances.
4. Payroll and payroll taxes: on the actual payroll calendar, in the actual weeks it lands. This is the line the whole forecast exists to protect.
5. Benefits: health insurance, retirement contributions, on their real schedule.
6. Rent and utilities: in the weeks they clear, not smoothed monthly averages.
7. Software and technology: EHR, telehealth, phones.
8. Billing and processing fees: your billing service or processor’s cut.
9. Loan payments: every note and line of credit, on its real schedule.
10. Owner pay and everything else: draws, plus a catch-all for the rest.

Each week ends with a projected balance: starting balance plus cash in, minus cash out. That projected balance becomes the next week’s starting point, so the columns chain into a single line you can read left to right. Where that line dips is your future, dated.

Why 13 weeks: it is one quarter. Long enough to contain every payroll cycle, every monthly bill, and the full collections lag, so a building problem has room to surface. Short enough that your estimates stay honest. You can estimate the next quarter from your own history; estimating next year week by week is fiction.

The estimates do not need to be precise. A typical collections week, the real payroll number, the real rent. Rough is fine. A forecast that is 90 percent right still shows a dangerous week a month early, and that warning is the entire product.

The ten-minute weekly routine

The forecast earns its keep through a short weekly habit, not through the initial build:

1. Close the week. When a week ends, mark it as actual and type in the real ending bank balance from your account. Everything downstream recalculates from reality instead of from your guess.
2. Read the difference. Compare the real balance to what the forecast projected. A small gap is noise. A large gap is information: a payer paid late, a bill you forgot, a deposit you double-counted. Adjust the coming weeks if it changes the picture.
3. Extend one week. Add your estimates to one more column at the far end, so you are always looking about 13 weeks ahead.

Step two is where the value compounds. The first month, your projections will miss by embarrassing amounts, and each miss teaches you something specific about your own practice: which payer actually runs 40 days, which “monthly” bill really clears on the 4th. After six or eight weeks of closing against actuals, most owners can project a normal week within a few thousand dollars. That skill stays even if the spreadsheet doesn’t.

The one-payroll-run rule

You need a threshold that tells you when a projected week deserves attention. The working floor I use is one payroll run: wages plus employer payroll taxes for one pay period. Any week where the projected balance dips below that number gets flagged. Below zero is a different category entirely.

To be clear about the ideal: one payroll run is a floor, not a target. A practice with one to two months of operating expenses in reserve rides out a slow payer without drama, and that is the position worth building toward. But many therapy and ABA practices live closer to the floor than anyone would design on purpose, and a threshold you will actually monitor beats a target you have already given up on.

When a week flags, work the options in order of speed:

1. Collections first. Pull your AR aging and chase the oldest claims and unworked denials. This is your own money arriving faster, and it is usually the largest lever. (If you don’t currently watch your AR aging, the billing KPIs piece covers what to pull.)
2. Timing second. Move discretionary outflows, owner draws and non-urgent purchases, out of the thin weeks. The forecast shows you exactly which weeks can absorb them.
3. Credit last, but early. If a line of credit might be needed, arrange it while your balance is still healthy. Lenders price desperation, and the difference between asking six weeks early and asking the week of payroll is the difference between a rate conversation and a rescue.

All three options depend on lead time. That is the real product of the forecast: not precision, weeks of warning.

What this forecast won’t do

A 13-week forecast is not a budget. It makes no judgment about whether your spending is right; it only projects timing. It is also not an accounting document. It runs on cash in and out of the bank, ignores accrual questions entirely, and will not reconcile to your P&L.

It is only as good as your knowledge of your own payer timing, which means the first month’s projections will be off, sometimes meaningfully. That is expected and self-correcting if you close the weeks against actuals. And it will not fix a margin problem. A practice that loses money on every session will see the loss coming with beautiful clarity, but seeing it is not solving it. If the forecast keeps flagging weeks and collections are not the issue, the problem is usually in the middle of the P&L, not in the timing.

Get the template

I built a free version of the forecast I use with clients: a 13-week cash flow template for therapy and ABA practices, in Excel with a Google Sheets option. It has the ten lines above, the forecast-to-actual weekly routine built in, automatic warnings on any week that dips below one payroll run, and an example practice filled in so you can see a summer dip play out before you enter your own numbers.

Download the 13-week cash flow template. Setup is about 15 minutes, and the weekly habit is about ten. If you set it up and something in your numbers doesn’t make sense, that is a reasonable thing to bring to an intro call.