Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Glossary term

Demand Pull-Forward

Purchases a promotion moves earlier in time rather than creates, as customers who would have bought later at full price buy during the offer.
BusinessconceptUpdated

Definition

Pull-forward is the shift of purchases earlier in time caused by a promotion, as opposed to purchases the promotion created. A customer who would have bought next month at full price and buys this weekend at a discount adds an order to the sale, removes one from next month and gives away the discount on a sale the business would have made anyway. The guide to measuring a sale or promotion describes it as demand borrowed from the future, and adds the mirror image: when customers expect a sale, they delay purchases, and the weeks before it fall below normal too.

The effect is largest where customers buy the same thing repeatedly and can predict their own need: consumables, replenishment products and anything bought on a regular cycle. A promotion aimed at regular customers is the most likely to borrow from their next order.

In practice: reading the weeks before and after a promotion

Pull-forward cannot be seen in the promotion's own numbers. It appears only against a baseline, an estimate of what sales would have been with no promotion, built from comparable past periods or from a group of customers who were eligible but not sent the offer. The measuring guide reads actual against baseline orders across three windows: before, during and after. During the promotion the difference is positive; before and after it is usually negative. The net across all three is the gain.

The length of the window after the sale matters. It should be set from how often customers normally buy, so that the people who bought early have had time to miss their usual next order. Stopping the reading on the promotion's last day is what makes pulled-forward demand look like growth. The discount depth test builds a follow-up window into its design for this reason, and the email frequency test warns that a short reading period rewards the higher cadence for sales that would have happened anyway. The guide to planning promotions brings the same warning forward to the planning stage, so the measurement window is agreed before the offer goes out.

Pull-forward, stock-up and cannibalisation compared

Pull-forward is one of three effects that together explain most of the gap between a promotion's headline orders and its real gain.

EffectWhat happensWhere it shows
Pull-forwardOrders move from later weeks into the promotionSales below baseline in the weeks after, and sometimes before
Stock-upCustomers buy more units per order, then return later than usualUnits per order during the sale and a longer gap to the next order
CannibalisationDiscounted sales replace full-price ones at the same timePromoted against unpromoted products, and codes used by people already buying

Pull-forward and stock-up are both timing effects; cannibalisation happens in the same moment. The distinction matters because each is read in a different place, and a review that checks only one leaves the others in the headline.

Illustrative example — not a real result

A candle maker reads the weeks after its weekend sale

Ambervane Candle Works normally takes about 300 orders a week, roughly 90 of them at weekends. A 25% weekend sale brings 180 orders in two days, double the usual weekend, and the team reports 90 extra orders. Its baseline, built from the same weeks in the two previous years, expects 300 orders in each of the following three weeks. Actual full-price orders come in at 270, 275 and 285, a shortfall of 70. The net gain is 20 orders, not 90, and each of the 160 sale orders that would have happened anyway carried the discount. The next sale goes only to customers who have not bought for six months.

Incrementality is the question pull-forward complicates: the incremental gain is what the promotion added after borrowed and cannibalised sales are taken out. Seasonality has to be in the baseline, because a dip after a sale can be the season turning rather than demand that was borrowed. Contribution Margin decides whether the net gain was worth it, since every pulled-forward order carries the discount. A Holdout Group of eligible customers who did not receive the offer gives the cleanest baseline against which pull-forward can be measured.

Common confusions

  • A post-sale dip means the sale hurt the business. The dip is borrowed demand coming due; the question is whether the net across all the weeks was positive after margin.
  • Pull-forward and cannibalisation are the same. Cannibalisation replaces a full-price sale at the same time; pull-forward moves one from later.
  • Platform reports show it. Ad platforms report orders during the campaign. The guide to sales and seasonal peaks on Meta sends that question to the business's own order data, read against a baseline.
  • Only the weeks after matter. Sales expected on a regular date can depress the weeks before as well.
  • Bigger peaks need bigger budgets. Budget for a peak is set from expected incremental gain, not the headline; the marketing budget calculator for seasonal peaks helps plan spend around those periods.

More terms to look up

Calculators

Where this term is used

Related