Thirty minutes from me, ninety from you. I want you to argue with this, not agree with it.
Friday 21 August 2026. Every figure comes from the board-approved FY26 budget and our payroll file. If one looks wrong, say so while I am on that slide.
The problem
That is $1.9 million a year. Nothing we spend money on today is big enough to absorb it.
Why the money is going away
Three of those four were always going to end on those dates. They were in the budget the board approved in March. Our grant income drops from $2.87 million to $1.02 million, which is a fall of 65 percent.
What we did when the grants ended
| Who pays each dollar of salary | January to April | October to December | Change |
|---|---|---|---|
| A grant pays for it | 2,595,195 | 1,103,545 | -1,491,650 |
| Clinical work we bill for | 924,635 | 924,635 | no change at all |
| Overhead, which has no income behind it | 626,897 | 2,118,547 | +1,491,650 |
One and a half million dollars of salary moved from a grant paying for it to nobody paying for it. The amount we bill for did not move by a single dollar.
That is the real problem. Cutting jobs is necessary, but cutting jobs is not what fixes this.
The plan, and the assumptions behind it
Each bar is one of the four things I will walk through next. The nine positions and the two consultants who have already left are fixed, so they are not sliders. If you set clinical income to 80 percent of budget, the monthly gap gets $57,000 worse. That is why that slider is first.
Saving money, 1 of 4
A grant pays for the work, not for a person. So when we cut someone a grant was paying for, we move someone from overhead into that slot. The grant pays for them instead, and our own costs drop.
There are three limits on this, so we should not promise more than it can do. The person has to do work that genuinely qualifies under that grant. Each grant caps how much salary we can charge to it without asking permission first. And we have to keep records of who spent time on what.
Saving money, 2 of 4
That split decides who we tell first and who takes over their work.
We are not cutting a service. We are buying it instead of employing someone to do it.
Before any of this happens I need a name against every client. Five of the nine carry a caseload and no document we have records how many clients each one has. That is a query Erin can run, and I want it before we tell anybody.
Saving money, 3 of 4
| What we spend it on | Per year | What we spend it on | Per year |
|---|---|---|---|
| Jobs we budgeted for and never filled | 489,250 | Subscriptions and software | 54,000 |
| Billing done by an outside firm | 160,250 | Computer equipment | 54,000 |
| Accounting done by an outside firm | 150,000 | Insurance, all four policies | 179,215 |
| Recruitment | 56,000 | Cleaning | 65,000 |
| Other consultants | 54,000 | Phone and internet | 27,000 |
Be careful with the first line. Not filling a job we budgeted for helps the budget, but it does not put cash in the bank, because that money was never going out in the first place. The amount that actually reaches our bank account is closer to $444,000.
One item I want to raise on its own. We pay $94,800 a year in interest, which means we owe somewhere around $1.2 million. I want to go and talk to that lender before they come and talk to us.
Saving money, 4 of 4
I cannot do this part from a spreadsheet. I need Ariel and the clinical team to own it.
The question that actually decides it
This uses the sliders from the previous slide, plus the $77,000 of wind-down money that arrives up front. Savings build up over time: the jobs go on 1 September, expense cuts land over the first month, clinical improvement takes ninety days.
This is the part I want you to sit with. Even if we do everything on the last four slides, including the $42,000 a month from clinical work, we still do not reach 1 December on cash. Two things change that. If the SAMHSA replacement lands we finish around $141,000 to the good. If it does not, we need roughly $100,000 of bridge to get there.
So there is a third job alongside cutting costs and billing more: find the bridge, or find out about SAMHSA sooner.
A monthly gap tells us the shape of the problem. This tells us whether we get to the decision.
The new structure
We should do this even if we were not short of money. Running these as separate departments is why psychiatry patients do not reliably get to therapy and why case management is underused. We wrote that down in the stabilization plan before we counted a single job.
Dates
| Date | What happens |
|---|---|
| 31 Aug | Three things on the same day: the new structure, the changes to how services work together, and the expense cuts. If we do the first without the second, the savings never appear and we will have put people out of work for nothing. |
| 1 Sept | The new structure starts. |
| 30 Sept | The SAMHSA grant ends. It is our largest. |
| Sept to Nov | We track cash payroll by payroll. Where we are at sixty days tells us what ninety days will look like, which leaves thirty days to change course. |
| 1 Dec | We decide whether to merge, wind down, close, or carry on. If we do the work above, that is a decision we get to make. If we do not, someone else makes it for us. |
This is a separate decision on a separate timeline, and it is the only thing here that the board has to approve.
The policy we have now
Someone who leaves in their second month has already been given twelve months of leave. We carry that as money we owe, and we pay it out when they go.
The policy I am proposing
| Years working here | What they get today | What they would get | Change |
|---|---|---|---|
| One to two | 10 vacation plus 5 personal, so 15 days | 10 days | 5 days fewer |
| Three | 15 vacation plus 5 personal, so 20 days | 15 days | 5 days fewer |
| Four to five | 20 vacation plus 5 personal, so 25 days | 15 days | 10 days fewer |
| Six and over | 20 vacation plus 5 personal, so 25 days | 20 days | 5 days fewer |
| Sick leave | 10 days | 10 days | no change |
Everyone loses at least a week. People in their fourth and fifth year lose two. I think it is right, but it is not a small change and we should not call it one.
What the change is worth
| What the change does | If everyone loses five days | How reliable is that number |
|---|---|---|
| Reduces what we owe in unused leave, and what we pay out when someone leaves | 63,927 | Reliable, but it only turns into cash when somebody actually leaves. |
| Frees up clinician days we could bill for | 68,297 | I do not believe this one. Our own stabilization plan says several clinicians do not have enough clients to fill the days they already have. |
| Reduces payroll | 0 | We pay salaried staff the same whether they take the day or not. |
I am not going to present this as a cost cut, because it is not one. If I did, the first question would be why payroll did not go down, and I would not have a good answer.
The honest reason to do it is that it reduces what we owe and brings us in line with how comparable agencies run their leave. That is a good enough reason on its own.
How it gets approved
One more thing we cannot decide on our own. Pausing the retirement contribution usually needs a formal change to the plan document and written notice to staff. We should ask before we name a date.
Anything we cannot settle today goes on the wall. Every source behind these numbers is in the full working document.