Promotional lift is the increase in units or sales during a promotion over the baseline you would have sold without it, usually stated as a percentage of that baseline. Incremental sales are what remains of the lift after you subtract sales pulled forward from the weeks after the promotion and sales taken from your own sibling products.
Lift is the number that shows up in the recap. Incremental sales are the number that decides whether the promotion paid. The two can differ widely, as the example below shows, and the gap is invisible in a daily sales chart because the chart shows only the promotion week rising.
Three things eat lift. Baseline is the obvious one: the units that would have sold anyway now sell at a discount, and every one of them is margin handed back. Pull-forward is the second: shoppers who would have bought next week buy this week, and the week after the promotion goes quiet. Cannibalization is the third: a shopper who would have bought the 12-pack buys the discounted 6-pack instead.
Platform reports do not separate these for you. Amazon's ad attribution, as described on advertising.amazon.com, credits a Sponsored Products seller campaign with purchases within 7 days of a click on a last-click basis, and reports "halo" sales of other products from the same seller alongside the advertised product. Attributed sales measure who clicked before buying. They do not measure who would have bought without the ad, so attributed and incremental are different questions.
Illustrative numbers. Take a seller with a kitchen product that normally sells 60 units a day at $30.00. Contribution after all fees, ad cost, and COGS is $9.00 per unit at full price. The seller runs a five-day promotion at $24.00. On Amazon, a $6.00 price cut also cuts the 15 percent referral fee by $0.90, so contribution falls by $5.10 to $3.90 per unit.
What happened:
Now the decomposition.
The recap says 133 percent lift. The net figure is 195 units across 12 days.
Now profit. Over the 12-day window the focal product earned 700 times $3.90, which is $2,730, plus 315 times $9.00, which is $2,835, for $5,565. Without the promotion it would have sold 60 a day for 12 days, 720 units at $9.00, or $6,480. The focal product is down $915. The sibling lost $800 of contribution.
The promotion generated real incremental units and still lost money. Split the 700 promotion units by where they came from and the focal product's $915 shortfall reconciles exactly:
Sum: minus $915. Add the sibling's $800 and the promotion is $1,715 behind doing nothing.
Measuring lift needs a clean baseline period and a clean promotion period, at the SKU level, with fees and refunds attached to the right dates. The comparison report sets two periods side by side by product, which is where baseline and promotion weeks get compared.
Crunch, the AI CFO inside ConnectBooks, takes questions in the pricing investigations area: whether a price change made money, or what pausing ads on a product did to profit. On each answer it finds what changed, compares the right periods, finds the products responsible, explains why, and recommends a next step. Asked about the promotion above, it can compare the promotion window against the weeks before and after, and show the sibling SKU's decline in the same period.
What it cannot do is prove causation. It sees that the sibling fell during the promotion. It does not know whether a competitor also cut price that week, whether a listing was suppressed, or whether the sibling went out of stock in one warehouse. A baseline is an estimate, and an AI that reads your books inherits every flaw in the estimate. Treat the incremental figure as a reasoned estimate you check, not a measured fact.
The weeks just before Black Friday are a poor baseline because shoppers hold off ahead of the event. A better baseline is the same SKU's sales in the matching week last year, scaled by the change in your non-promoted sales since then. If the SKU is new, use the average of three to four non-promoted weeks and accept that the estimate is rough.
Both are common. Units are cleaner for the decomposition, because price changes during a promotion distort dollar lift. Profit decisions need contribution dollars, as in the example above.
At least as long as the promotion itself, and longer for products bought on a replenishment cycle. Consumables can show a dip for several weeks.
Sometimes. It can be worth running to clear aged stock, to earn reviews on a new product, or to rank a listing. Those are legitimate reasons. Write the reason down before the promotion starts, so the loss is a known cost of a stated goal rather than a surprise in January.
Running an e-commerce business comes with plenty of challenges, but ConnectBooks is here to make your life easier. With real-time insights, seamless integrations, and detailed tracking of your profitability and inventory, you can stay ahead of the game. Whether you’re selling on Amazon, Shopify, Walmart, TikTok or eBay, ConnectBooks helps you manage your finances with 100% accuracy and confidence, so you can focus on growing your business.
Ready to level up? Start making smarter, data-driven decisions every step of the way. Try ConnectBooks Free Today or Schedule a Demo
