Every spring, K-12 leadership teams produce an enrollment forecast that determines next year's staffing model, facilities plan, and budget envelope. A forecast that is 3% high quietly creates a hiring problem; a forecast that is 3% low becomes a class-size and operations problem in September. The teams that get this right do not have better intuition — they combine historical patterns with current pipeline signals and document the assumptions so the forecast can be challenged before it becomes a plan.
Pipeline signals to weight, not guess
A credible forecast starts from observable pipeline data, not from a hunch about whether next year "feels strong." That pipeline typically includes: open inquiries received year-to-date, applications submitted, accepted offers, deposits or registration confirmations, and re-enrollment intent from current families. Each stage has a historical conversion rate to the next, and the right tool to track that flow is the same student tracking system that will eventually hold the student record. Borderset keeps applicant and enrolled records in one place, so a "what does the pipeline look like compared to this date last year" question takes a single view rather than a registrar export.
Attrition modeling that doesn't pretend
Most forecasts get the new-students half right and the attrition half wrong. Attrition is rarely uniform — it concentrates in grade transitions (5 to 6, 8 to 9), in families who moved during the school year, and sometimes in specific bus routes or programs. Model it by cohort using the last three to five years of actual moves, not by a single percentage applied flat across the school. When the cohort model and a flat-rate model disagree by more than a couple of points, the cohort model is almost always closer.
Scenario planning, not a single number
Leadership decisions are easier when the forecast is a range with named scenarios — base, downside, stretch — each tied to specific assumptions about re-enrollment rate and new-student conversion. Borderset's enrollment view lets leaders toggle assumptions and see staffing implications side by side. That converts "how confident are you?" from a feeling into a comparison.
Two case patterns worth borrowing
Multi-campus groups that grow quickly tend to forecast best when each campus runs its own pipeline view and the central office aggregates only at the cohort level. The Level Up case study walks through how that played out as the group scaled from 2 to 9 campuses without a forecast meltdown. Larger single-site schools face a different challenge — pipeline volume is high enough that re-enrollment intent dominates the forecast, which is the pattern the British school case study describes.
The forecast itself rolls into the principal dashboard so the staffing and facilities teams see the same numbers the board reviews — without the version-mismatch problem that kills credibility.
What to stop forecasting
Drop the single-line forecast that contains no assumptions. Replace it with: a scenario range with stated re-enrollment and conversion rates, a grade-by-grade attrition view drawn from the last three years, and a weekly tracker of pipeline progression vs. the same week last year. Those three artifacts let a superintendent challenge the forecast in the meeting where it matters — before staffing offers go out and facilities decisions are locked. Borderset is built so all three live in the same workspace as the eventual student record, which removes the worst part of forecast season: reconciling tools.
Enrollment forecasting is not a prediction exercise — it is a planning exercise. The right tools make the assumptions visible enough that the plan survives contact with September.