What cannibalization is
Cannibalization is when a new SKU, flavor, or promotion steals volume from your own existing items rather than pulling it from a competitor. I learned to fear the word the quarter my brand launched a single-serve version of our bestselling 12oz oat granola at Sprouts. The single-serve "grew" 6,000 units in its first 12 weeks, the launch deck looked triumphant, and the category buyer was thrilled. Then I pulled the full lineup and the 12oz had quietly dropped 5,400 units over the same stretch. We had moved volume from one shelf tag to another and called it growth.
If you typed "cannibalization meaning" into a search bar after a launch review where one item soared and the total stayed flat, this is the gap you were staring at. The new thing is up, the category is not, and nobody in the room wants to say why.
Gross lift vs. net lift
Here is the distinction that the launch deck almost never makes. Gross lift is what the new item sold. Net lift is what the new item added to the total after you subtract whatever it pulled off your other SKUs. Internal cannibalization is the difference between the two.
Run the granola numbers across a 12-week post-launch window in SPINS-shaped form:
| SKU | Pre-launch units | Post-launch units | Change |
|---|---|---|---|
| Oat granola 12oz | 41,000 | 35,600 | -5,400 |
| Oat granola single-serve | 0 | 6,000 | +6,000 |
| Almond granola 12oz | 18,000 | 17,800 | -200 |
| Brand total | 59,000 | 59,400 | +400 |
The single-serve's gross lift is 6,000 units. The net lift to the brand is 400 units. Cannibalization ate 5,600 of the 6,000, a 93% cannibalization rate. The new SKU is real, the shoppers buying it are real, but they are overwhelmingly my own shoppers trading down to a smaller pack, not new households. That is a very different launch story than the one in the deck.
The same arithmetic catches promotional cannibalization. When you run a deep temporary price reduction on the 12oz, some of the "lift" is just your full-price 12oz buyers pulling forward purchases they would have made anyway, plus single-serve buyers switching to the discounted big pack. Separating the genuinely incremental units from the borrowed ones is the whole game, and it is the same separation our post-promo lift and baseline methodology walks through line by line.
When cannibalization is fine and when it is not
Cannibalization is not automatically bad. The question is always: did the trade come with a better margin, a defended facing, or a blocked competitor?
A few cases where I happily accept it:
- Margin upgrade. If shoppers trade from a $3.49 SKU to a $4.99 SKU at a similar cost to produce, cannibalized units can still grow contribution dollars even when unit count stays flat.
- Defensive launch. If a private-label competitor is about to take the single-serve slot at Kroger, launching your own single-serve to hold the planogram space can be worth heavy cannibalization.
- Pack migration you control. Moving buyers to a subscription multipack you sell direct can be a deliberate trade.
Where it stings is the case my granola launch fell into: net new revenue near zero, an extra SKU now consuming a slot, working capital, and a slotting fee, and a 12oz that looks like it is declining on its own when it is really just being eaten from the inside. That is exactly the kind of low-net-contribution item a disciplined SKU rationalization review is built to flag before it clutters the shelf for two more years.
Why analysts have to measure it
The reason cannibalization slips through is that most launch reporting looks at the new item in isolation. You pull the one SKU, you see a number going up, you stop. Measuring net lift means pulling the entire affected sub-segment, the new item plus every sibling that could plausibly lose volume to it, and watching the total. If the total moves less than the new item's gross sales, the difference came out of your own pocket.
This matters most at category-review time. When a Kroger or Whole Foods buyer asks whether your new item earned its keep, "it sold 6,000 units" is the wrong answer. "It added 400 net units and a margin point" is the right one, even when it is less flattering, because the buyer is running the same category-total math from the other side of the table. Tying launch decisions to net category impact is the core of real category management, not item-level cheerleading.
Where Scout fits
The slow part of measuring cannibalization is assembling the before-and-after picture across every affected SKU in a segment, which usually means hand-stitching SPINS pulls in a spreadsheet for each launch. Scout sits on your SPINS or retailer data and lets you compare a SKU's gross movement against the segment total over the same window, so a launch's net lift is a question you can answer instead of a spreadsheet you dread building. It measures the trade; it does not decide for you whether the trade was worth it.
The short version
- Cannibalization is volume a new SKU or promo pulls from your own existing items instead of from competitors.
- Gross lift is what the new item sold; net lift subtracts what it cannibalized. The granola single-serve had 6,000 gross and 400 net, a 93% cannibalization rate.
- It is not always bad. Accept it for a margin upgrade or a defended facing; kill the launch when net contribution is near zero and you are just paying to move volume between your own shelf tags.