A useful sales forecast for a small producer is a table: each product down the side, each month across the top, in units first and dollars second. The quickest honest way to fill it is to start from last year's months, which carry the shape of your year, apply a growth rate for each product, and add the new accounts you already know about. The forecast then drives everything else: what to make and when, what to buy and how far ahead, and how much cash the busy season ties up before it pays. In the example below, a cocoa maker forecasts $123,300 of sales, 62% of it in November and December, and works back to order its tins in September. Below is how to build it, a worked example, what to use it for, how to keep it honest, and a free template.
The short version
- Products down the side, months across the top, units first.
- Start from last year's units by month; that's your season.
- Apply a growth rate per product, not one for the whole business.
- Add new accounts you know about, as units a month.
- Turn units into what to make, what to buy and when, and the cash it takes.
- Each month, replace the past month with actual sales and adjust the growth.
- TaroStack fills forecasts from your own sales history, on Standard. More below.
Building it
Wasatch Cocoa Co., a made-up cocoa mix maker in Park City, Utah, I use in these examples, starts from last year:
| Product | Last year, units | Growth | New accounts, a month | Price |
|---|---|---|---|---|
| Classic cocoa mix, 12 oz tin | 2,630 | 15% | 30 | $16 |
| Peppermint cocoa mix, 12 oz tin | 1,485 | 25% | 0 | $16 |
| Vanilla marshmallows, 8 oz bag | 1,200 | 10% | 10 | $9 |
| Cocoa gift set | 535 | 30% | 0 | $38 |
Each month's forecast is last year's month × (1 + growth), plus the new accounts. The growth rates differ because the products do: peppermint and the gift set grew fastest last year, and the classic mix gains a new café account worth 30 tins a month. A product with no history borrows the shape of a similar one, scaled to a guess.
The forecast
| Product | Jan | Jun | Oct | Nov | Dec | Year |
|---|---|---|---|---|---|---|
| Classic cocoa mix | 375 | 76 | 329 | 628 | 927 | 3,384 |
| Peppermint cocoa mix | 100 | 6 | 188 | 525 | 875 | 1,857 |
| Vanilla marshmallows | 120 | 32 | 142 | 296 | 428 | 1,440 |
| Cocoa gift set | 13 | 0 | 39 | 195 | 416 | 697 |
| Sales | $9,174 | $1,600 | $11,032 | $28,522 | $48,492 | $123,302 |
November and December are $77,014 of the year, 62%; June is $1,600. A forecast that spread the year evenly would say $10,275 a month and be wrong every month. How to plan stock for a year shaped like this is in seasonal inventory.
What the forecast is for
What to make. December's 2,646 units can't all be made in December. With a kitchen that packs about 1,200 tins and bags a month, the plan starts building stock in September, so the forecast becomes a production schedule.
What to buy, and when. Printed tins take eight weeks to arrive, so November's and December's tins are ordered in September; cocoa powder and sugar on shorter lead times follow the production schedule. How far ahead to reorder is in reorder point example.
The cash. Ingredients, tins and hours for the stock built in September and October are paid before November's sales come in, which is the money a forecast tells you to have ready; how many sales a month cover the fixed costs is in break-even point for a food business.
Keeping it honest
At the end of each month, replace that month's forecast with what you sold, and compare: if October came in 20% over, November probably will too, and the growth rate goes up before the tins run out. A forecast that's never compared with actual sales is a hope, not a plan. For a product line where most sales come through a few large accounts, ask the accounts for their own forecasts and use those instead.
Get the template
Download the sales forecast template. Two sheets: last year's units by product and month; and the forecast, with each product's price, growth rate and new-account units a month, giving this year's units by month, the year's units and sales, and monthly totals in units and dollars. It's an ordinary Excel file with no macros, and it works in Google Sheets, LibreOffice and Numbers.
Where this stops working
Twenty products, last year's numbers dug out of a sales report, a forecast that has to become a production schedule and a list of what to buy, and a new product with no history. A sheet can't turn the forecast into how many tins to order and when, against your kitchen's real capacity.
How TaroStack does it
TaroStack fills forecasts from your own sales history, on Standard.
Forecasts from your history. Planning fills forecasts from your own sales history, so last year's months are already there, and you adjust them.
What to make and buy, and when. Planning works out what to make and what to buy, and when, against your capacity and each supplier's lead time, as drafts you approve, so the tins are ordered in September without anyone remembering to.
And the actuals. Sales orders record what really sold, margins are reported by product and by customer, and the reorder list works in days of cover, so the forecast and the shelf are compared every week.
| The job | Spreadsheet | In TaroStack |
|---|---|---|
| Last year's units by month | Dug out of a report | Your sales history |
| How many tins to make each month | Worked out by hand | Planning: what to make, when |
| When to order printed tins | Remembered, hopefully | Planning: what to buy, by lead time |
| Forecast against actual | Rarely checked | Sales by product, every week |
Recipes, batches and sales orders are on every plan, from $49 a month, and planning, forecasts included, is on Standard at $99. The first 30 days are free, with no credit card. If you'd like a hand setting up, ask, and we'll do it with you.
