Forecasting the renewal wave — then staffing for it

Anyone who runs a licensing, permitting or compliance desk knows the shape of the year: long stretches of steady work, then a renewal surge that swamps the team and blows the service-level targets. Two questions decide whether that surge hurts. How big is the wave — and how many reviewers does it take to clear? Vastvic answers both, and hands the second answer straight out of the first.

Forecasting the volume alone isn’t enough: two hundred renewals is a quiet week or a crisis depending on how fast your desk processes one and how many people are on it. So the pipeline runs end to end — forecast the renewal volume, then size the desk that absorbs it.

Knowing the wave is coming isn’t the answer. Knowing its size — and how many hands it takes — is.

Forecast the wave, two ways

How you forecast renewal volume depends on the data you hold, so the model ships in two modes:

• Portfolio projection. When you hold the active book with expiry dates, the wave is largely deterministic. The model buckets future expiries by period and category, applies a per-category renewal probability (your historical renew-versus-lapse rate), and returns expected renewals per week or month — with a binomial confidence band, widest exactly where the renew/lapse split is most uncertain.
• Historical-pattern forecast. When you only have past renewal counts, it projects them with a classic additive decomposition: a least-squares trend, seasonal indices that preserve the annual rhythm, and a residual band. No black box — closed-form and fully reproducible.

Both are deterministic: the same inputs always produce the same forecast, so a staffing plan is auditable rather than a number that changes every time you re-run it.

Then size the desk — it’s a queue

A renewal desk is structurally a queue: work arrives at some rate, each item takes some time to review, and a pool of reviewers works it off. That’s the exact problem classical queueing theory — the same Erlang-C mathematics behind call-centre staffing — was built to solve. The forecast’s peak period feeds straight into it, turning an expected renewal load into the numbers a manager plans against:

• Offered load and utilisation — how hard the current desk will be pushed.
• p90 wait time — how long the slowest 10% of renewals will sit, which is usually what the SLA is written against.
• SLA-breach risk — whether the current headcount holds under the wave.
• Recommended reviewers — the exact headcount needed to bring wait times back inside target.

Run it across renewal categories — by licence class, region, or vendor — and the surge stops being a surprise and becomes a staffing plan with a number on it, weeks out.

---

The useful forecast for a compliance team was never just “how many.” It’s “how many renewals, when — and therefore how many people.” Forecast the wave, model the desk as the queue it is, and the annual renewal scramble turns into capacity you can schedule.