Incrementality is the share of sales, units, or profit that happened only because of a specific action, such as an ad campaign, a promotion, or a new sales channel, measured against what would have happened without it. Incremental lift is that difference expressed against the baseline, and an incrementality test is any comparison built to estimate the baseline honestly, usually with a control group that did not get the action.
Every platform reports its own contribution generously. An ad console counts sales that followed a click. A new channel's dashboard counts every order it took. Neither asks whether those buyers would have bought from you anyway, through organic search, your own store, or another marketplace. Add the dashboards together and a multichannel seller can account for more sales than the business made.
Incrementality is the correction. It moves the question from "what did this channel or campaign touch" to "what did it add." Spend and effort should follow the second number. Contribution margin on attributed sales can say a campaign is profitable while contribution on incremental sales says it loses money, and only the second one shows up in the bank.
Amazon's own metrics illustrate the distinction. Its ad campaign attribution page on advertising.amazon.com defines new-to-brand purchases as purchases by customers who had not bought from the brand in the previous 365 days. That is a useful acquisition signal. It is not incrementality: a new-to-brand buyer who clicked an ad might still have found the product organically.
Illustrative numbers. Take a seller of pour-over coffee kits that sells on Shopify and adds TikTok Shop in October. Contribution per unit is $13.20 on Shopify and $8.40 on TikTok Shop, where affiliate commissions and platform fees take a larger cut. A separate product line sold only on Shopify, and never listed on TikTok Shop, serves as the control.
The baseline. With no launch, Shopify kit sales would have been expected to move with the control: 1,000 times 0.98 equals 980.
Cannibalized units. 980 minus 820 equals 160 Shopify kits that went to TikTok Shop instead.
Incremental units. 400 TikTok Shop units minus 160 cannibalized equals 240. Sixty percent of the new channel's orders were new to the business.
Incremental contribution. TikTok Shop earned 400 times $8.40, or $3,360. Shopify lost 160 times $13.20, or $2,112. Net: $3,360 minus $2,112 equals $1,248.
The TikTok Shop dashboard shows $3,360 of contribution. The business gained $1,248, or $3.12 for each TikTok Shop order. The channel still adds money. It adds about 37 percent of what its own numbers suggest, and every buyer who moves from Shopify to TikTok Shop costs $4.80 in margin. Past about 255 cannibalized units the channel loses money: at 280, the Shopify loss is 280 times $13.20, or $3,696, more than the $3,360 TikTok Shop earned.
Promotions follow the same logic, with pull-forward from later weeks as an extra subtraction; the promotion version is worked in promotional lift. The ad version, with a pause test and a control group, is in separating incremental sales from sales you would have had anyway.
An incrementality comparison needs each channel's sales and contribution by product and period in one ledger, so a decline in one channel can be set against a gain in another. ConnectBooks reconciles TikTok Shop, Shopify, Amazon, Walmart, eBay, and Temu sales into QuickBooks or Xero with SKU-level profit, which is the input this kind of question needs.
Crunch, the AI CFO inside ConnectBooks, takes period-comparison questions on the seller's reconciled ConnectBooks data: what changed, which periods to compare, which products moved, why, and what to do next. Asked "did coffee kit profit change after October, and which products moved," it can set the two periods side by side and find the products whose sales shifted.
It cannot establish causation on its own. A Shopify decline that coincides with a launch could come from an email that did not go out, a stockout, or a site change. The control group is your judgment, and so is the conclusion.
Lift usually means the raw increase over baseline during an action. Incrementality goes further and nets out what came from elsewhere, such as sales pulled from later weeks or taken from your own channels. Lift can be large while incrementality is small.
Roughly. Compare the period before and after the action, then adjust with a control group that did not get it. That is weaker than a randomized test, because the control is never a perfect match, but it is far closer to the truth than attributed sales.
Long enough to cover at least one full weekly cycle on each side, and usually two. Short tests miss delayed effects such as repeat purchases and organic ranking changes. Run a second test at a different time before making a permanent cut.
Yes. A new marketplace can grow its own sales while shrinking another's. Measuring the net change across all channels is the only way to know what the launch added.
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