The key idea

Observed sales can be limited by supply. If a book is unavailable, the absence of completed sales does not prove that no customers wanted it. Record stockout periods alongside sales history and treat them as constrained observations. Any estimate of missed demand should remain an estimate, not become invented transaction data.

This is an original operational explainer. Its examples and decision rules are illustrations, not measured customer results or universal service commitments.

Track the constraint at the right level

Availability can differ by warehouse, edition, seller account and market. A paperback unavailable in one route does not establish a worldwide shortage for the title. Record when the relevant offer could actually be fulfilled, and distinguish physical stock from sellable availability. A mapping failure or a blocked listing may restrict sales even when books are on the shelf. The forecast needs the constraint affecting its own target order stream.

Separate observed and imputed demand

Several modeling approaches can reduce the influence of constrained periods or estimate what might have happened. Whatever approach is chosen, retain the original measured orders and flag any inferred values. Evaluate the assumptions on periods where demand and availability were observable. For an operational pilot, conservative handling with manual review can be preferable to a precise but untested missed sales estimate. Explain the uncertainty before it changes a production quantity.

Worked example · illustrative assumptions

Consider a fictional title selling five copies a day before running out for three days. The warehouse then receives stock and sales resume. Adding fifteen lost copies to the history is a hypothesis based on the earlier rate, not a fact. A planner might use that scenario to assess a safety allowance, but should show the measured zero fulfilled sales and the constrained period separately rather than quietly replacing the records.

Put it into practice

  • Record availability constraints by edition and fulfillment route.
  • Keep observed orders separate from inferred missed demand.
  • Review replenishment decisions that depend heavily on stockout assumptions.

Watch out for

Do not reward a forecasting model for predicting low sales during a known shortage if the business goal is to anticipate unconstrained demand. The target and evaluation must match the inventory question.

Is every zero sales day evidence of a stockout?

No. A fully available book can have no orders. Use actual availability records where possible rather than guessing a constraint whenever sales are disappointing.

AI assisted editorial content with original planning examples. Source links support the stated source notes; operational suggestions are guidance rather than guarantees. Provider rules and product coverage can change. Editorial approach · BooksAhead.ai connection status.