A Simple Sales Forecast: Projecting Revenue From Past Growth

The sales history

Monthly sales for a small business over six months:

Sales rose in every month, by about 4% to 6% each time. The goal is to estimate months 7 to 9.

Method 1: straight-line trend

Assume sales keep growing by the same amount each month. Find the average monthly increase, then add it repeatedly.

average monthly change = (latest − first) ÷ number of intervalsSix data points make five intervals.

Here the change is (25,300 − 20,000) ÷ 5 = 1,060 a month. Month 7 is estimated at 25,300 + 1,060 = 26,360, and month 9 at 25,300 + 3 × 1,060 = 28,480. This method suits steady, mature businesses where growth is roughly a fixed amount each period.

Method 2: compound growth rate

If growth is closer to a constant percentage, each month builds on a larger base. The average monthly rate across the data is the geometric one:

monthly growth rate = (latest ÷ first)^(1 ÷ intervals) − 1

That gives (25,300 ÷ 20,000)^(1/5) − 1 = about 4.81% a month. Month 7 is 25,300 × 1.0481 = about 26,518, and month 9 is 25,300 × 1.0481³ = about 29,132. The same calculation is performed by a CAGR calculator when you enter the first and last values and the number of periods.

Compounding pulls ahead of the straight line over time: by month 9 the two methods differ by about 650. Over a year or two, the gap becomes large. Small monthly rates add up faster than intuition suggests: 3% a month for 12 months is 1.03¹² = 1.4258, a 42.6% rise over the year, not 36%.

Method 3: moving average

When sales bounce around without a clear direction, averaging the last few months smooths out the noise. A three-month moving average of months 4, 5 and 6 is (23,100 + 24,000 + 25,300) ÷ 3 = 24,133.

That is below the latest month, because it deliberately lags. It is a sound baseline for a flat or seasonal business, but it will under-forecast during steady growth, as it does here.

The three methods side by side

The first two agree closely over a short horizon, and the moving average sits well below because it assumes no growth. The spread is useful in itself: it is a sensible range to plan around. You might order stock for the low figure and keep the option to top up toward the high one.

For a business with strong seasons, a fourth approach is to take the same month last year and scale it by recent yearly growth. If last December sold 31,000 and sales are up 12% on a year ago, the December forecast is 31,000 × 1.12 = 34,720.

Choosing a method and checking it

• Look at the pattern first. Roughly constant increases point to a straight line, constant percentages to compound growth, and no trend to a moving average.
• Allow for seasonality. If December is always strong, compare each month with the same month a year earlier and apply the yearly growth rate (see the percentage increase calculator) to it.
• Give a range. Run a cautious, a likely and an optimistic case. Planning on the cautious one keeps surprises small.
• Keep horizons short. Forecasts from a handful of data points deteriorate quickly beyond a few months.
• Compare forecasts with actuals. Each month, note the gap and adjust. Consistent over- or under-shooting tells you which assumption is wrong.

This article is general education about simple arithmetic, not financial or business advice. Forecasts are estimates and real results can differ, so avoid committing money you cannot afford to lose on a projection.