How small businesses should forecast sales (without becoming a data scientist)

Why forecasting matters for SMBs

Large enterprises can absorb forecasting errors. Small businesses can't. A 20% over-forecast can mean over-hiring, over-ordering inventory, and a cash crunch 90 days later. A 20% under-forecast can mean stock-outs, missed revenue, and lost customers. Both scenarios are preventable with even basic forecasting discipline.

Most small business owners don't forecast at all, or use a single number ("last year + 10%") that breaks the moment anything changes — a new product launches, a competitor opens nearby, the economy shifts. This guide gives you three forecasting methods that work for businesses that don't have a finance team.

Method 1: Rolling trailing average (the simplest)

For a steady-state business with limited seasonality:

• Compute the average daily revenue over the trailing 90 days.
• Multiply by the number of days you're forecasting.
• Adjust for any known events (planned promotion, holiday, new product launch).

This takes 30 minutes to set up in a spreadsheet and is genuinely accurate for stable businesses. Most B2B services and recurring-revenue businesses fit this profile.

Method 2: Seasonally-adjusted trailing (most ecommerce + retail)

For businesses with meaningful seasonality (most retail and ecommerce):

• Compute year-over-year growth rate from the same period last year.
• Apply that growth rate to last year's revenue for the forecast period.
• Adjust for known disruptions (supply chain issues, promo plans, competitor activity).

For businesses with at least 18 months of history, this method outperforms most paid forecasting tools.

Method 3: Pipeline-driven (B2B with sales process)

For any B2B business with a defined sales pipeline:

• Sum the expected value of each pipeline opportunity, weighted by historical close rate at its current stage.
• Add committed recurring revenue (renewals, recurring subscriptions).
• Subtract expected churn based on at-risk accounts.

This method is sensitive to assumptions about close rates and stage definitions. The most common mistake is using close rates that are aspirational rather than historical — which leads to systematic over-forecasting.

The five forecasting mistakes

1. Ignoring seasonality. A 90-day average projected forward is useless if you do 40% of your annual business in December.
2. Using gross instead of net. Forecast net revenue (after refunds, churn, contraction), not gross.
3. Using aspirational close rates. Pipeline forecasts that assume "this time we'll close 50%" when historical is 25% are decisions disguised as forecasts.
4. Single-point forecasts. Always forecast a range: best case, base case, worst case. Plan to the base case, prepare for the worst.
5. Not updating. A forecast that's six weeks old is fiction. Re-forecast at least monthly.

Tools that do this for you

For pure spreadsheet-based forecasting, Google Sheets or Excel with the FORECAST.ETS function (which handles seasonality automatically) is genuinely good and free.

For automated forecasting with cross-tool data integration, Illuminated Intelligence [blocked] builds rolling forecasts from your Stripe [blocked], Shopify [blocked], QuickBooks [blocked], and CRM data automatically. The platform produces three-scenario forecasts (best/base/worst) updated daily, and JARVIS surfaces why the forecast moved when it does — "Pipeline coverage dropped because three enterprise deals slipped to next quarter."

Forecasting is one area where automation genuinely outperforms manual work because the data sources and assumptions need to update constantly. Manual forecasts get stale fast; automated ones don't.

Ready to see your business, illuminated? Start a free 14-day trial [blocked] of Illuminated Intelligence — no credit card required, full setup in under an hour. Or meet JARVIS [blocked], our AI business advisor that turns your data into next-step recommendations.