How to define a category in retail

Why the boundary is the first call

$3.3M of the Sprouts refrigerated salsa set hangs on one category definition call: whether the refrigerated queso and dips sharing the cold fixture, 18% of it, belong in the salsa category or a neighboring one. Include them and the set is $18.4M. Exclude them and it is $15.1M, and Verde Fresca's dollar share jumps almost a full point without a single extra unit selling. Nobody voted on that $3.3M swing. Whoever drew the boundary decided it, usually on the first slide, usually without saying so out loud. I run the numbers both ways before I walk in, because that one line moves the total category size, my brand's share, and which segment gets called the growth engine.

Defining the category is step 1 of the eight-step category management process, and it is the most consequential call precisely because nothing downstream sees the decision being made. The assessment, the scorecard, the share math, the space-to-sales read on the planogram: all of them just inherit whatever universe you drew at the start and report on it as if it were fixed. A buyer who accepts a supplier's category definition without checking it has let that supplier pre-load every number in the deck. So this page is how to draw the line on purpose, the way a shopper actually shops the set, and how a good buyer catches you if you draw it to flatter your own brand.

Rack the way the shopper substitutes, not the way the warehouse ships

The warehouse racks refrigerated salsa by how it moves through the supply chain: temperature zone, case pack, distributor, planogram bay. None of that is how a shopper decides what to buy. The method that holds up in a review is the consumer decision tree: define the category the way a shopper actually substitutes and segments the set, by the need they are filling, the form they reach for, the occasion they are buying for. If a shopper reaching for fermented probiotic salsa would sooner walk out empty-handed than swap in a jar of shelf-stable pico, those two things are not competing for the same purchase, and a definition that lumps them together is describing the shelf, not the demand.

The tree is easiest to build by running the set through a few segmentation lenses at once and seeing where the natural breaks fall. Here is the Sprouts refrigerated salsa set through five of them:

Read across the rows and the fault lines show up. Fresh red salsa, fermented salsa, and salsa verde all pass the substitution test at the edges: a shopper who wanted chunky pico might grab restaurant-style instead, and a verde buyer might trade to a mild red on a given trip. Queso is the row that does not behave. On need-state it is a warm cheese indulgence, not a fresh-produce snack. On occasion it skews party and game-day, not weekly restock. On ingredient it is dairy, not tomato. Three of the five lenses put queso on the far side of a real substitution boundary. That is the signal that it is a neighboring category riding along in the same cold fixture, not a segment of this one. This is the same demand-first logic behind defining category roles, which is the very next step and inherits whatever universe you settle here.

Wide versus narrow, and the supplier's thumb on the scale

Once you accept that the boundary is a judgment call, you have to be honest that suppliers have a direction they want it to go, and it is not always the same direction. A wide definition makes the category look bigger and the growth story grander, which helps if you are pitching the buyer that the whole set deserves more shelf. A narrow definition shrinks the denominator, which flatters your own share and helps if you are arguing your brand is a bigger deal than it looks. Whichever way your interests point, you will feel the pull, so the discipline is to name the trade-off out loud rather than pretend the line is objective.

The tell a buyer looks for is simple: does the person proposing the boundary happen to look better under it? When Casa Marisol, the category captain at ~29% share, proposes a wide definition, the buyer should ask whether the extra size comes from segments Casa Marisol actually sells into or from queso lines it barely touches. When I propose a definition that spotlights fermented salsa, where two of Verde Fresca's five SKUs live and are growing 34% YoY, the buyer should and does ask whether I have drawn the category down to the one room where my brand is winning. Both moves are legitimate to make and legitimate to catch. The category validator, the #2 supplier who checks the captain's work, exists largely to catch the first one.

The rule I hold myself to: define the category to describe the shopper honestly, then argue my brand's case inside that honest definition. A boundary drawn to win the argument gets found out the first time the buyer runs the segment math herself, and then every other number in my deck is suspect too.

A worked example: the Sprouts refrigerated salsa set, both ways

Here is the full set at Sprouts, roughly 345 stores, on the two candidate definitions. The only thing that changes between them is whether refrigerated queso and dips is in or out.

Now pull queso out. The salsa-only universe is $18.4M minus the $3.3M queso segment, which is $15.1M. Every remaining segment keeps its dollars but takes a bigger slice of the smaller pie:

Now watch what it does to my brand. Verde Fresca sells $1.42M at Sprouts. In the wide $18.4M category that is a 7.7% dollar share, good for #4 in the set. In the narrow $15.1M salsa-only category the same $1.42M is 9.4% share. Same brand, same five SKUs, same $42 per store per week velocity against a category median of $31. Nothing about the business changed. I moved the boundary and picked up 1.7 points of share, enough to change how the #4 brand gets talked about in the room.

The growth story flips too. The wide set is up 6.2% in dollars, and a chunk of that headline is queso doing +7%, a segment I do not compete in at all. Strip queso out and the salsa-only set grows a little slower on the top line, but the number I actually own, fermented at +28%, becomes a much larger share of the story it tells. If I want the buyer to fund a fermented expansion, the narrow definition is the one that makes my case, and that is exactly why she will pressure-test it.

The honest answer here, the one I would actually recommend to the buyer: queso and dips is a neighboring category. It fails the substitution test on need-state, occasion, and base ingredient, and it belongs in its own review with its own captain. So the working category is the $15.1M salsa-only set, and Verde Fresca is a 9.4%-share #4 brand in it. I will say that out loud, show the buyer the wide number too so she knows I am not hiding it, and then argue fermented on the merits. That is a definition I can defend when Tolteca Foods, the validator, re-runs it.

How category definition feeds everything downstream

The reason to spend real time on step 1 is that steps 2 through 8 cannot reopen it without redoing all their own work. Two examples make the dependency concrete.

Definition feeds role. Whether refrigerated salsa is a routine category or gets read as something with more pull depends partly on what is inside it. Fold in party-occasion queso and the blended set skews more toward entertaining and impulse, which nudges the category role and, with it, how much shelf and promotion the category can argue for. Keep the set to fresh salsa that shoppers restock weekly and it reads as a cleaner routine category. The role decision is only as sound as the boundary it sits on.

Definition feeds assessment. The category assessment, where the analyst's real work lives, is entirely a function of who is in the set. Include queso and the set grows 6.2%; exclude it and the salsa-only growth rate, segment mix, and every share figure shift. The famous assessment line for this category, up 6.2% in dollars but only 1.1% in units, so roughly five points of the growth is price and not real demand, is itself boundary-dependent: queso and salsa do not have the same price-versus-volume split, so mixing them blends two different stories into one average. Get the boundary wrong and the most important sentence in the whole review is measuring the wrong universe.

This is why I never treat the definition as boilerplate to clear before the "real" analysis starts. It is the analysis. Every metric in the deck is a statement about a set of products, and the set is something a person chose.

Doing this in Scout

Scout does not draw the boundary for you, and it should not. Deciding whether queso and dips belongs in the salsa category is a judgment about how Sprouts shoppers actually substitute, and that judgment is the buyer's and the captain's to make and defend. What Scout does is make the two-universe comparison a two-minute check instead of a half-day of rebuilding pivots. You define a category as a set of items or segments once, save it as a view, and then toggle a segment in or out to see total category dollars, each segment's share, and your own brand's share recompute against both boundaries side by side. That is the difference between the $18.4M wide read and the $15.1M narrow read, with Verde Fresca at 7.7% or 9.4%, sitting next to each other so you can argue the honest one on purpose rather than defaulting into whichever pivot you happened to build first. Scout assembles the evidence for the boundary. The person in the room still owns the line.

The short version

• The category boundary is step 1 and the most consequential call, because every downstream number, assessment, scorecard, share, and shelf, inherits the universe you draw without re-examining it.
• Draw the line by the consumer decision tree (how shoppers substitute across need-state, form, and occasion), which is why queso and dips, at 18% of the Sprouts fixture, reads as a neighboring category rather than a salsa segment.
• Wide definitions inflate category size and dilute your share; narrow ones flatter it (Verde Fresca moves from 7.7% to 9.4% just by excluding queso), so define the shopper honestly first, then argue your brand inside that line.

Related: How to run a category assessment · Category roles: destination to convenience