How to forecast demand for a subscription product

Split your demand into two streams

Subscription products usually have two sources of demand: scheduled recurring shipments to existing subscribers, and one-off purchases from new customers or gift buyers. Mixing them into one average sales number gives you a distorted forecast, especially if the one-off orders are lumpy or seasonal.

Forecast the recurring stream using your subscriber count and ship schedule. If you have 400 active subscribers on a monthly plan shipping on the 1st, you need 400 units ready by that date, regardless of what daily sales looked like last month. Forecast the one-off stream using normal sales history and trend, the same way you would for a non-subscription SKU.

Watch churn and plan changes

Subscriber counts move over time. People pause, cancel, upgrade, or switch variants. Before each shipment cycle, check your current active count rather than relying on last cycle's number, especially if you are growing or running a promotion that adds a batch of new subscribers at once.

If subscribers can change quantity or frequency, track the net effect on units per cycle, not just headcount. Ten subscribers switching from monthly to every-two-months cuts that stream's demand in half even though your subscriber count looks unchanged.

Build in a buffer for timing risk

Subscription shipments are often batched and inflexible on date. If stock is short on ship day you cannot push the order into next week without disrupting your schedule or annoying customers. Carry extra safety stock for the recurring stream, more than you would for a product where backorders are tolerable.

Step by step

1. List your active subscriber count per variant and confirm it before each ship cycle, not once a quarter.
2. Calculate recurring unit demand as active subscribers x units per shipment for the upcoming cycle.
3. Pull your one-off (non-subscription) sales history separately and forecast it using normal trend and recent averages.
4. Add the two streams together to get total expected demand for the period.
5. Build safety stock on the recurring stream large enough to cover a late supplier shipment without missing ship day.
6. Recheck subscriber counts a few days before the order-by date so last-minute cancellations or signups are reflected.
7. Adjust quantities immediately after any promotion that adds or removes a batch of subscribers at once.

Worked example

Say you sell a monthly coffee subscription with 350 active subscribers getting one bag each, shipping on the 5th of each month. Recurring demand for next cycle is 350 bags. Non-subscription sales of the same bag have averaged 60 units over the last 28 days, roughly 2.1 units a day, so over the 25 days until the next ship date that adds about 53 units. Total expected demand is 350 + 53 = 403 bags. If lead time plus safety days is 20 days and daily sales for safety stock purposes is treated as 350/30 = 11.7 units a day for the recurring stream, you would want roughly 234 units of safety stock on top of the base need, so order enough to cover 403 units plus that buffer, minus what you already have on hand and on order.

How SteadyShelf helps

SteadyShelf forecasts from your actual sales history in Shopify, so it will pick up the recurring orders generated by your subscription app as part of that sales data, but it does not read subscriber counts or ship schedules directly. For accurate recurring-stream forecasting, calculate expected subscriber shipments yourself and treat SteadyShelf's suggested quantity as the baseline for one-off demand, adding your subscriber number on top before approving the purchase order.

$29/month flat, unlimited products and suppliers, 30-day free trial. Run the numbers now with the free reorder point calculator.

SteadyShelf is launching on the Shopify App Store. Join early access to get the install link first.

Related guides

• How to forecast demand for accessories that sell with a main product
• How to forecast demand for products with sizes and colors
• How to forecast demand for slow moving products
• How to forecast demand for a product that keeps selling out

All inventory how-to guides · Inventory glossary

FAQ