Forecasting Demand From Last Year's Numbers

Why this matters

You cannot staff, stock, or schedule a season you have not predicted. Most field-service shops already own the data they need to forecast next season: last year's job records sit in their system, unused. A rough forecast built from real history beats gut feel every time, and it does not take a spreadsheet wizard to build. This is how you turn the numbers you already have into a month-by-month plan for hiring, parts, and cash.

Step 1: pull last year's job data by month

Start with what actually happened. Open your records and count, month by month, for the last year (more if you have it).

• Total jobs completed each month.
• Total revenue each month.
• The service mix: which job types drove each month.

If you have two or three years, all the better. Multiple years separate a real seasonal pattern from a one-off fluke. One bad-weather year can distort a single-year view.

Write the monthly numbers in a simple table. The shape of the year jumps out the moment you see it laid out.

Step 2: find the seasonal pattern

With the monthly numbers in front of you, identify the repeating shape of your year.

• Mark your peak months, your slow months, and the shoulder weeks between them.
• Note when demand starts climbing and when it falls off. The transitions matter more than the peaks.
• Calculate each month as a rough share of the year, so you can see how lopsided your business is.

This pattern is the backbone of every seasonal decision: when to launch marketing, when to ramp staff, when to stock parts, when to brace for the slump.

Step 3: adjust for what was unusual

Last year was not a clean experiment. Before you project it forward, strip out the noise.

• Weather events: a freak storm or a mild season distorts the count. Note it and discount it.
• One-off jobs: a single large project that will not repeat should not set your baseline.
• Capacity limits: if you turned away work because you were full, real demand was higher than your completed-jobs count shows. Account for the lost work, or you will under-forecast.

The goal is a "normal year" baseline, not a literal copy of a year that had its own quirks.

Step 4: layer in what changed this year

History is the base; this year's known differences are the adjustment.

• Growth or shrinkage: if you are bigger this year (more crew, more marketing), scale the baseline up. If smaller, down.
• Market shifts: new competition, a major local employer change, or a shift in your customer base.
• Your own plans: a new service line or a heavier pre-season campaign will change the curve you inherited.

Apply these as adjustments to the baseline, not replacements for it. The history anchors you; the adjustments tune it.

Step 5: turn the forecast into operational decisions

A forecast that just predicts numbers is useless. Convert it into the decisions it should drive.

• Staffing: size your seasonal ramp and your overtime plan to the forecast peak. (See related: The Seasonal Staffing Ramp-and-Cut Decision Tree.)
• Cash: project the lean months and confirm your reserve covers fixed costs through them.
• Parts and inventory: stock the high-demand items ahead of the months that consume them.
• Marketing: time the pre-season push to land work in the soft weeks before the forecast climb.

Each prediction should point at a concrete action. If it does not change a decision, you do not need to forecast it.

Step 6: track actuals against the forecast and correct

A forecast is a hypothesis. The value compounds when you check it against reality and learn.

• As the season runs, compare actual jobs and revenue to what you predicted.
• Note where you were off and why (weather, a missed market shift, a capacity ceiling).
• Feed the corrections into next year's forecast so it gets sharper every cycle.

After a few years of this, your forecast becomes genuinely reliable, and the seasonal scramble fades into a planned routine.

References

• SBA (U.S. Small Business Administration): demand forecasting and business-planning guidance for small businesses.
• IRS: recordkeeping practices that make historical job and revenue data usable.
• See related: Seasonal Cash Management; The Seasonal Staffing Ramp-and-Cut Decision Tree.
• See related: The Seasonal Marketing Calendar.