Why the category assessment sets the whole agenda
The $18.4M number on the front of a Sprouts refrigerated salsa deck is the least useful thing in a category assessment. It is big, it is true, and it decides nothing. The set was up 6.2% in dollars last year, which sounds like a category to invest behind until you put the 1.1% unit line next to it and see that roughly five points of that growth was price, not shoppers buying more salsa. A buyer who runs her own numbers finds that gap in about four seconds, and the "category is booming" story built on the dollar headline is dead before the second slide. The dollar figure is not the assessment. The pairing that reframes it is.
That is the thing about the category assessment, step 3 of the eight-step process. It is where a brand analyst's leverage actually concentrates, more than any single item recommendation. The assessment is the read of the category, and the read sets the agenda for the entire review. Steps 5 through 8, strategy, tactics, implementation, review, all inherit from it. Get the read wrong and every downstream decision aims at the wrong problem: you defend shelf for a segment that is dying, or you cut a SKU that was the only thing growing. Get it right, in the metrics the buyer already trusts, and you have already written the first draft of her plan before she has opened hers.
The six reads, and the trap in each one
An assessment is a small set of reads that only mean something sitting next to each other. Here is the framework I run every time, on the $18.4M Sprouts salsa set or the Kroger broth set or anything else. Each row is a read, what it tells you, and the trap that turns it into a wrong answer.
| Read | What it tells you | The trap |
|---|---|---|
| 1. Dollars vs. units | Whether growth is real demand or just price and mix | Reporting the dollar line alone. +6.2% dollars looks like a win until you see +1.1% units under it |
| 2. Dollar and unit share trend | Whether the category is gaining or losing ground against itself and the market | Confusing a share gain from one competitor's stumble with real category momentum |
| 3. Velocity per store per week vs. category median | The quality of distribution, not its width | Reading velocity with no benchmark. $42 means nothing until you know the category median is $31 |
| 4. Distribution: ACV and TDP | How much room the category or item has left to expand | Reading an ACV move without checking store count moved with it |
| 5. Segment mix shift | Where the growth engine actually sits inside the category | Averaging the segments together and missing that one small segment is carrying the whole set |
| 6. Gap analysis | Where the category is leaking to another channel, segment, or price tier | Stopping at "we grew" and never asking where the demand that left the shelf went |
None of these is exotic. The discipline is refusing to report any one of them alone. A dollar figure with no unit figure, a velocity with no median, an ACV move with no store count: each of those is a sentence with the verb missing, and a buyer who runs her own numbers will finish it for you in a way you do not control.
Read units next to dollars, always
If you take one habit from this page, take this one. The single most important move in any category assessment is reading units directly next to dollars, in the same row, every time.
The Sprouts salsa set was up 6.2% in dollars and 1.1% in units last year. Subtract, and about five points of that growth is price and mix, not more jars leaving the shelf. That gap is the whole story. A plan that treats 6.2% as demand over-invests in a category whose real unit velocity is nearly flat, and it will miss on volume even if it hits on dollars. The same discipline is channel-agnostic. A shelf-stable broth set at Kroger might read +4% dollars and -1% units in 84.51 data, an even starker version of the same thing: dollars up, units actually down, pure price with real demand eroding underneath it. Different channel, different data provider, identical read.
Here is why I put them side by side rather than reporting a blended "growth" number:
| Set | Dollars YoY | Units YoY | What the gap says |
|---|---|---|---|
| Sprouts refrigerated salsa | +6.2% | +1.1% | ~5 pts is price/mix; real demand nearly flat |
| Kroger shelf-stable broth | +4% | -1% | ~5 pts is price; unit demand is eroding |
Dollars measure what the register rang. Units measure whether people wanted more of the thing. When those two diverge, the divergence is the finding, and it decides whether the category strategy should be a growth push or a margin-and-defense hold. For the fuller map of which metric answers which question, the SPINS metrics decision tree is the companion to this page. But the dollars-vs-units read is the one I would never let leave the slide unpaired.
A worked example: the Sprouts refrigerated salsa set
Let me run the whole framework end to end on the set I know best, the $18.4M refrigerated salsa and dips category at Sprouts, roughly 345 stores, off SPINS. This is the part to read if you skip everything else.
Read 1, dollars vs. units. The set is +6.2% dollars, +1.1% units. Already flagged: growth is mostly price. Note it and carry it into every read that follows, because it colors what "up" means everywhere else.
Read 5, segment mix. Before I look at any single brand, I break the category into its segments, because the aggregate hides the engine:
| Segment | Share of category $ | YoY $ | Read |
|---|---|---|---|
| Fresh refrigerated salsa | 62% | +3% | The big, mature base. Growth roughly tracks price, so real demand is soft |
| Fermented / probiotic salsa | 11% | +28% | Small but the actual growth engine. This is where units are moving |
| Salsa verde / tomatillo | 9% | +4% | Steady, unremarkable |
| Refrigerated queso & dips | 18% | +7% | The adjacency question: does this belong in the set at all? |
The mix read reframes the category. Fresh is 62% of dollars and barely moving on units. Fermented is only 11% of dollars but growing 28%, and given the category's flat unit line, fermented is where nearly all the real incremental demand is coming from. If you averaged the segments together you would call the category "up mid-single-digits" and completely miss that one 11% slice is carrying it.
Read 3, velocity vs. median. Now my own brand, Verde Fresca: $1.42M at Sprouts, 7.7% dollar share, #4 in the category, 5 SKUs. Its items average $42 per store per week against a category median of $31 per item. That is a 35% premium to the median. In the velocity read, a number well above the median means the item works hard in every store that carries it. The product is not the problem.
Read 4, distribution: ACV and TDP. Verde Fresca sits at 78% ACV at Sprouts. So the item does $42 a week where it is stocked, comfortably above median, but it is missing from stores worth 22% of the chain's volume. Hold those two facts next to each other: strong velocity, incomplete distribution. That is the shape of an under-distributed winner. Every point of ACV I can recover is landing on a per-store rate that already beats the field. For the mechanics of ACV and total distribution points, what is ACV and what is TDP cover the definitions I am leaning on here.
Read 2, share trend, plus the SKU cut. Verde Fresca holds 7.7% dollar share, but the average hides a split. Two of the five SKUs are the fermented line, growing +34% YoY, riding the same wave that lifted the whole fermented segment 28%. The other three fresh SKUs are flat, tracking the soft fresh segment. So my brand's growth is entirely the fermented pair. That matters for what I ask for: I do not want more shelf for five SKUs evenly, I want it concentrated where the units are moving.
Read 6, the gap. Put it together. The category's real growth engine is fermented, not the fresh base that fills most of the shelf. Demand in fresh is soft, its dollar growth is basically price. And Verde Fresca, the #4 brand, is under-distributed relative to how well it velocities, at 78% ACV on a $42 rate against a $31 median. The gap worth acting on is not "sell more salsa." It is: the shelf is over-weighted to a flat fresh segment while a 28%-growth fermented segment is under-spaced, and a proven fermented performer is missing from a fifth of the chain. That is a specific, defensible hole, and it happens to be one my fermented line is built to fill. The buyer can check every number, which is exactly why it works.
The four questions a good assessment answers
Strip the reads down and an assessment is done when it can answer four questions in plain language. If your slide cannot, you have a pile of metrics, not an assessment.
| Question | The read that answers it | The Sprouts salsa answer |
|---|---|---|
| Where is the growth coming from? | Dollars vs. units, plus segment mix | Fermented salsa: 11% of dollars, +28%, carrying a category whose units are otherwise flat |
| Which way is share moving? | Dollar and unit share trend | Fresh base holding on price; fermented gaining real unit share |
| Is there distribution headroom? | ACV and TDP vs. velocity | Yes. Verde Fresca at 78% ACV with a $42 velocity has room to add doors that will perform |
| What is the biggest gap? | Gap analysis across the above | Shelf over-indexed to flat fresh; a high-velocity fermented item under-distributed |
Notice the answers build on each other. Growth source tells you fermented is the engine. Share direction confirms it is real unit share, not a price mirage. Distribution headroom says the fix is reachable, there are doors to gain. And the gap ties them into one sentence a buyer can act on. That last sentence is the deliverable. Everything upstream exists to earn the right to say it.
Anti-patterns that produce confident wrong answers
Most weak assessments I have seen do not fail on math. They fail on a habit that looks fine until someone pushes back in the meeting.
Leading with dollar growth only. The one that got me on my first deck. "+6.2%" is a true number and a useless strategy input if units are flat under it. Any assessment that opens on a dollar figure with no unit figure beside it is one buyer question away from collapsing. Pair them in the same row, every time, and let the gap be the headline.
Not benchmarking velocity to the category. A $42 per store per week velocity sounds like a fact until you ask, against what. Against a $31 category median it is a 35% premium and an argument for more doors. Against a $60 median it would be a below-average item you should be careful pitching for expansion. The raw velocity is meaningless. The ratio to the category median is the read. Never present one without the other.
Reading ACV moves without checking store count. ACV is volume-weighted, so it can shift on panel composition or a single large-volume account changing hands, with no actual change in how many stores carry the item. If ACV ticks from 78% to 81%, I do not call it distribution growth until I confirm the store count moved in the same direction. An ACV gain with a flat store count is usually a weighting artifact, and reporting it as new distribution is how you promise a buyer doors that were never gained.
The through-line: every one of these is reporting a number without the number that makes it mean something. Dollars without units, velocity without a median, ACV without store count. The assessment is the pairing, not the metric.
Doing this in Scout
The grind in an assessment is assembling it. To get the six reads above you are pulling your items and the competing items, this year against last, dollars and units side by side, ACV and velocity per store per week, then rebuilding the segment mix by hand, out of raw SPINS or Circana exports. Then you do it again six months later for the next review. It is the same pivot every cycle, which is exactly the kind of recurring build that eats an analyst's week.
Scout sits on that data and holds the assessment as a saved view: dollars and units paired, velocity against the category median, ACV and TDP, and the segment breakout, refreshed off the latest extract instead of rebuilt from scratch. The two-minute version is refreshing last quarter's assessment before a review. The half-day version, slicing the fermented segment by region to see whether the 28% growth is national or a West Coast story, is still real work, but you start from an assembled read instead of a blank pivot. What Scout does not do is make the call. It helps you bring the evidence. The buyer still owns the read of what to do about the gap, the margin target, and the shelf, and the strategy that follows the assessment is step 5, still a human decision. Scout gets you to a defensible assessment faster. It does not run the review for you.
The short version
- The assessment is step 3, and it is where the analyst's leverage concentrates, because the read of the category sets the agenda for the whole review and every later step inherits from it.
- Read units next to dollars, velocity next to the category median, and ACV next to store count, because each metric is only a finding when it is paired with the one that gives it meaning.
- A finished assessment answers four questions, growth source, share direction, distribution headroom, and the biggest gap, and on the Sprouts salsa set the gap is a flat fresh segment over-holding shelf while a high-velocity fermented winner sits at 78% ACV.
Related: Building a category scorecard · The 8-step category management process