Enrolment is up this year. The intake pool peaked last year
How do we plan for the demographic cliff while enrolment is still growing?
Figures and rules on this page apply to
United States
Working somewhere else? The shape of the problem usually travels. The deadlines do not.
What this looks like
The spring numbers come in ahead of last year, so the pressure comes off. Nobody asks which programmes grew and which shrank, because the total went the right way. Two years later a department is unviable and the decision has to be made in one budget cycle instead of four.
The numbers
Every figure here is someone else’s. Check them.
- 3.9MUS high school graduates in 2025, the highest on record
- 3.4Mprojected in 2041, a 13 percent fall from the peak
- 38states projected to see fewer graduates in 2041 than in 2023
- -32%projected fall in Illinois, the steepest of the large states
- +1.3%actual undergraduate enrolment growth in spring 2026, to 15.5 million
- -8.4%fall in computer and information sciences enrolment at four-year institutions
Why it happens
It is not a people problem.
Two curves are moving in opposite directions at once. The pool of eighteen-year-olds has turned down, while participation and transfers have held current enrolment up. A single total cannot show both, so an institution reading one number sees good news. The mix underneath is where the decision lives, and the mix is exactly what no single system reports.
Why your current software has not fixed it
Because it was never built to.
The student information system is built to be the record of a student, and it is good at that. It is not built to answer a portfolio question. Which programme, at which campus, for which cohort, is moving which way against what it costs. That question needs admissions, registration, aid, cost and completion in the same frame, and those are four systems with four owners. The SIS vendor is aimed at the record, not at the portfolio, and that will not change.
Intelligence, plumbed in
An adviser's note predicts a withdrawal weeks before the data does.
The build is one view across admissions, aid and cost at programme level. Then a projection from your own feeder schools, and a Monday list. Underneath it: a connector, one agreed meaning per field, and a test set scored on your own records.
How we make AI survive real data-
Connectors -
A semantic layer -
Evals you can check
However hard, whatever it is
Enrolment is one example. Bring what the board asks about.
This is the pressure everyone can name. The ones that are harder to name, and closer, are usually where the decision actually sits.
-
01 We sit with you
Days where the work happens, not a workshop in a meeting room. We watch the job get done and write down the shortcuts nobody wrote down.
-
02 We read everything
Your data, your rules, your vendors and their documentation, and the published research on your sector. We report what is actually in there.
-
03 We break it to first principles
Not which tool fixes this. What is actually causing it, taken apart until we reach the piece that cannot be divided further.
-
04 Then we build
Weeks, not quarters. By this point we are not guessing what to build, and guessing is the thing that makes projects long.
What we build
Specific enough to argue with.
Four mechanisms, not four features. Each one is a thing that happens on its own, every day, whether or not anyone remembers to run it.
-
One view across admissions, registration, aid and cost, at programme level. The mix stays visible while the total is still reassuring.
-
A cohort projection that takes your own feeder-school data and the published graduate projections for your states, rather than a national average that describes nobody.
-
An early-warning list a tutor or adviser gets on a Monday, built from attendance, activity and financial holds together instead of each one separately.
-
A screen for the department office showing the week ahead. The people who can act on a signal are not the people who read dashboards.
How you would know it worked
Numbers in your own reporting, not ours.
- Programme-level contribution, stated as a range rather than a point, reviewed termly instead of annually.
- Share of at-risk students contacted within a week of the signal appearing.
- Forecast error on your own intake against actuals, which is the only honest test of a projection.
Straight answers
Where a model is involved, it is scored against your own records first. Accuracy per source, not one flattering average.
The questions this raises
-
Our enrolment is growing. Why would we act now?
Because the projections are about school leavers, not about you, and the lead time is the whole point. Acting while the totals are healthy is a choice; acting after they turn is a reaction with fewer options.
-
Do you write into Banner or SITS?
No. We read them. Nothing is written into a student record without an approved route. A wrong write on a student record is not recoverable the way a wrong report is.
-
We are outside the US. Do these numbers apply?
The projections do not. The shape does - the UK, much of the EU and South Korea face their own version with different dates. We would use your own published figures, and say plainly which ones we could not find.
-
Is this a dashboard project?
Only partly. A dashboard nobody opens is a cost. The useful half is the Monday list and the office screen, which put the signal in front of the person who can act on it.
Where the figures come from
We did not make these up, and you should not take our word for them.
Somebody noticed first, and nothing consumed it
The earliest warning is a note in a box nothing reads.
A driver, a fitter, an engineer or an adviser saw it coming and wrote it down. It went into free text. Free text does not trigger anything, does not add up, and never tells you that four people reported the same thing this week.
- ConstructionBuilding productivity rose. Highways and bridges went backwards
- Energy & utilitiesMajor events now cost more outage hours than every ordinary year combined
- ManufacturingProductivity fell in fourteen of the twenty manufacturing groups measured
- StaffingFive million people hired and five million separated in the same month
- Transport & fleetMore than a third of crashed trucks already had out-of-service defects
Different work, same box nobody reads. Whatever your people notice first, it is being written down somewhere already.Tell us what yours write down.
The total is reassuring. The mix is the decision.
The build is one view across admissions, aid and cost at programme level. Then a projection from your own feeder schools, and a Monday list.
See everything we build-
Software -
Hardware -
Ways of working -
Whole ventures
Is this happening to you? Tell us the size of it.
Twenty minutes. We will tell you honestly whether the numbers justify doing anything about it.