How do you forecast next month's service call volume well enough to plan around?
Start with the same weeks last year, adjust for how your business has actually grown since, then adjust again for known drivers: weather outlook, planned marketing, and maintenance visits already owed. Forecast at weekly resolution. You will not predict the exact number and you do not need to; you need the shape of the month and the week where capacity breaks.
A baseline anyone can build
The workable method is deliberately simple. Take last year's call volume for the same weeks. Multiply by your current year-over-year growth rate, measured on recent months rather than the full year. Add the maintenance visits you already owe in that window, because those are contractual demand you created yourself. Then adjust for anything structurally different: a new location, a territory change, a campaign that will not repeat.
This gets you close enough to plan staffing and marketing. Elaborate models rarely beat it by enough to matter at the resolution operations actually schedules at.
The drivers worth adding, in order of value
- Owed maintenance. Known demand sitting on your books. Most companies forget to include it and then wonder why the board filled up.
- Weather outlook. Temperature extremes drive emergency volume with a short lag. Even a rough seasonal outlook improves the shape of the forecast.
- Planned spend. Campaign launches, seasonal promotions and budget changes move volume on a predictable schedule.
- Backlog. Unsold estimates and deferred repairs represent demand that will surface, often on the first hot or cold day.
Forecast the shape, judge yourself on the breaks
A forecast that is off by a modest percentage on the monthly total but correctly identifies the week your schedule saturates has done its job. A forecast that nails the total and misses the timing has not, because staffing and spend decisions are made week by week.
So evaluate forecasts on the questions they were built to answer: did we know which week would break, and did we act early enough to matter? Keep a simple record of each forecast and what actually happened. After a few cycles the systematic errors become obvious and easy to correct.
Where the data lives
Call history comes from the phone platform, job and maintenance obligations from the field service system, spend from the ad platforms. Whether the operational side runs on ServiceTitan, Jobber or something else, the forecast needs those three streams joined on a common calendar. That join is the only technical work in the whole exercise.
Once it exists, the forecast can be refreshed automatically and delivered with the numbers that matter beside it, which is exactly the job of recurring intelligence briefs: put the week's expected demand next to current capacity in front of the person who can act on it.
Topics: forecasting · demand · capacity planning · weather
Have a version of this question about your own business?
The useful answer usually depends on which systems you run and how they're connected. That's a conversation, not a blog post.