The key idea
Intermittent demand contains many periods with no orders and occasional sales of varying size. A single average can hide both the gaps and the bursts. For these books, inventory policy should consider order timing, replenishment speed and the cost of holding a few copies, not only a smooth daily forecast.
The forecasting reference discusses practical cases including sparse demand. Stock choices and the example here are our own operational analysis. [1]
Separate order frequency from order size
A title that sells one copy every ten days is operationally different from a title that sells ten copies in one occasional event. Track the number of demand occasions and the quantity on each occasion. Verify that zero periods are genuinely covered and not stockouts or missing imports. Group comparable titles cautiously when individual histories are short. A large catalog can provide useful patterns, but unrelated formats and markets should not be merged just to increase the sample.
Consider whether stock is the right response
A very quiet title may be better served by production after ordering if the promised delivery experience permits it. Another title may justify one or two copies because its small physical footprint and customer value make immediate availability worthwhile. Make that a deliberate service policy. Review slow stock periodically without forcing every title into the same target. The system should be able to recommend no speculative replenishment, not only a smaller print quantity.
Worked example · illustrative assumptions
In a hypothetical twelve week period, a title sells two copies in week three and three in week ten, with all other weeks confirmed at zero. The average is about 0.42 copies per week, but no customer orders 0.42 of a book. A useful policy might compare holding one or two copies with waiting for an order, considering the actual replenishment route. Five copies sold in total does not justify five copies permanently on the shelf.
Put it into practice
- Record both order occasions and units per occasion.
- Verify zero demand periods before fitting a sparse history.
- Compare a small stock policy with production only after an order.
Watch out for
Do not interpret one bulk order as a new daily run rate. Determine whether it was a repeatable demand change or a specific event before increasing the standing target.
Should every backlist title have a safety stock?
No. Availability goals, unit economics and production timing should determine which titles merit stock. Some books can remain order produced without disappearing from the catalog.
Sources and further reading
- Hyndman and Athanasopoulos: Practical forecasting issues. Sparse counts, missing data and practical forecasting limitations.
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.