On-Shelf Availability Benchmarks: Kroger 2026

On-shelf availability is the gap nobody reports

Most launch decks track distribution as a single number: how many doors, what percent ACV. Almost none track whether the product is actually on the shelf in those doors on any given week. On-shelf availability, the share of store-weeks where a distributed item is physically present and buyable, is the number that decides whether distribution turns into sales. And in the data we pulled, it is worse than the deck assumes.

We looked at complete Kroger scan and shelf-condition data for six CPG brands on Scout, then computed the benchmarks below across roughly 58 million store-weeks of distribution. These are real numbers from a live pull, not estimates. Here is the one that stopped us:

The distribution was there. The shelf presence was not. That gap is lost sales that no distribution report captures, because distribution reports count where a product is authorized, not where it is available.

What this study covers

The panel is six consumer brands that sell through Kroger and share their Kroger point-of-sale and shelf-condition feeds with Scout. The window is each brand's most recent 13 weeks of 2026 data. The categories the panel touches include coffee and coffee mixers, shelf-stable and sparkling juice, craft and specialty soft drinks, energy drinks, hair and personal care, and pet food, so the benchmarks lean toward center-store grocery and health-and-beauty rather than fresh or frozen.

Two measures come out of it:

• Velocity, meaning dollars and units per store per week at the stores actually selling an item. This is the "how fast does it move once it is on the shelf" number.
• On-shelf availability, meaning how often a distributed item logs an empty-shelf event, measured against the base of store-weeks where the item was in distribution.

Both are computed from Kroger's own feeds: store-level scan data at the item, store, and week grain, and a shelf-condition feed reported at the item, division, and week grain. The methodology section below spells out the definitions, the sample, and the limits.

Velocity benchmarks: the median SKU and the spread

Pooling every qualifying item across the six brands (about 9,300 UPCs with at least four weeks of history in five or more stores), the median item sells about two units per store per week. In dollars:

The spread is the story. Distribution build-out gets all the attention, but a top-decile item moves nearly three times the volume of the median item on the same shelf in the same weeks. Winning more doors does not fix a rate-of-sale problem, and these numbers show how wide the rate-of-sale range runs inside a single retailer.

One composition note worth stating up front: a single broad health-and-beauty portfolio contributes about half the items in the pool, and its items skew low-velocity, so the pooled median leans that way. Weighting each brand equally instead lands the median closer to $20 per store per week. Either way, the shape of the distribution is the same: a long right tail, a modest median.

Velocity varies more by category than most brands assume

Rate of sale is not a brand trait so much as a category trait. Coffee and energy-drink items move very differently from hair-care items on the same store shelf:

Energy drinks are the clearest illustration of why a median alone misleads: the median energy-drink item does $19.62 per store per week, but the 90th percentile does $124.59, a six-fold gap driven by a handful of high-turn SKUs. A brand benchmarking its velocity against a category average would badly misjudge where it actually sits.

On-shelf availability: the leak in the funnel

Selling fast per store only matters at the stores where the item is actually on the shelf. Across the panel:

Put differently, for every 100 store-weeks a product spent in distribution, the shelf-condition feed flagged an empty shelf 26 times. Some of those are brief, some are recurring, but each one is a window where a shopper who wanted the product could not buy it.

The share of item-weeks touched by a gap is even more striking:

This is not a story about a few sloppy brands. Brand-level shelf-gap rates ranged from 12 to 33 empty-shelf events per 100 store-weeks, and the share of item-weeks with a gap ranged from 31% to 82%. Every brand in the panel had a meaningful availability problem. The variation is in degree, not in kind.

Some categories leak far more than others

The availability gap concentrates in specific categories, and the pattern is intuitive once you see it: heavy, fast-moving, or bulky-to-restock items go empty most:

In the panel, the worst-hit categories ran above 40 empty-shelf events per 100 store-weeks, with a gap in three quarters of their item-weeks. When an item empties that often, the lost sale is on the shelf, not on the shopper.

The DC is usually not the excuse

The instinct when shelf gaps show up is to blame supply: the warehouse must be out. The Kroger inventory feed argues otherwise. In the same weeks the shelf was showing empty, the distribution centers serving those stores were generally holding inventory. We are not publishing a precise "phantom out-of-stock" rate here, because matching a shelf event to a specific DC's on-hand position cleanly requires reconciling store-to-warehouse and item-to-case mappings that the two feeds do not share on a common key. But the directional finding holds: the shelf goes empty while upstream inventory exists. The failure is in the last hundred feet, at the shelf, not at the dock.

Why velocity and availability belong on the same page

These two benchmarks are usually owned by different teams and reported in different meetings. Sales tracks velocity. Supply or retail-execution tracks availability. Read together, they answer a question neither answers alone.

A distributed item doing $30 per store per week that goes empty 25% of its store-weeks is not a $30 item. It is a $30 item leaking a quarter of its shelf presence, and the recoverable sales from fixing availability almost always beat the cost of chasing incremental doors. This is the same lesson the distribution metrics teach from the other direction: ACV tells you the weighted quality of the doors you have won, and TDP folds door quality and assortment depth into one figure, but neither tells you whether the product is on the shelf this week. On-shelf availability is the measure that closes that loop.

If you want to put a number on your own distribution gap, the free Distribution Gap calculator runs the TDP math against category benchmarks with no login.

For a brand that already knows its distribution is strong, the highest return is usually not more doors. It is defending the shelf in the doors already won.

Methodology: how the numbers were built

Everything below is reproducible from one query against the brands' own Kroger feeds.

Panel. Six consumer packaged goods brands that sell through Kroger and route their Kroger point-of-sale and shelf-condition data through Scout. Test and internal tenants were excluded. The brands are kept anonymous; the numbers are aggregated so that no single brand is identifiable from the pooled figures.

Window. Each brand's most recent 13 weeks of 2026 data (a trailing quarter). Coverage end dates vary by brand from late winter to early summer 2026, so the figures describe recent performance rather than a single fixed calendar quarter.

Grain and sources. Kroger store-level scan data (item by store by week: scanned units, scanned retail dollars, store number, division, Kroger commodity) joined to Kroger's shelf-condition feed (item by division by week: empty-shelf event count) and cross-referenced against the Kroger DC inventory feed (warehouse on-hand by case code). All three are the retailer's own reporting, not modeled estimates.

Velocity definition. For each item, velocity is the total scanned volume divided by the number of distinct store-weeks in which the item sold, so it measures rate of sale where distributed, not diluted by stores that never carried the item. Dollars-per-store-week and units-per-store-week are computed the same way. Items qualify with at least four weeks of history in at least five stores inside the window, to keep thin, noisy items out of the percentiles. About 9,300 items qualify.

On-shelf availability definition. The shelf-condition feed is an event log: it records empty-shelf events, not a full presence-or-absence panel. To turn that into a rate, the events are measured against the distribution base from the scan data. Empty-shelf events per 100 store-weeks is total empty-shelf events divided by store-weeks of active distribution, times 100. Share of item-weeks with a gap is the count of item-by-division weeks with at least one empty-shelf event divided by all item-by-division weeks in distribution. The join key is the Kroger item code and division, which the scan and shelf feeds share. The pooled base is roughly 58 million store-weeks and 1.5 million item-by-division weeks.

What this is not. It is one retailer, Kroger, so it is not a cross-channel or "total US" read. It is six brands, weighted toward center-store grocery and health-and-beauty, so category mixes that lean fresh, frozen, or club are not represented. Because one broad portfolio contributes about half the items, pooled velocity leans toward its mix; the by-category cuts and the equal-weight median are provided so a reader can see past that. And because empty-shelf and DC-inventory feeds do not share a common item-and-location key, we report the shelf gap and the inventory position as directional companions, not as a single reconciled phantom-stockout rate.

Where Scout fits

Scout puts velocity, distribution, and on-shelf availability in one view off a brand's own retailer feeds, so the "how fast does it move" and "is it even on the shelf" questions sit side by side instead of in two different teams' spreadsheets. The benchmarks in this report were produced with the same pull any brand on the platform can run against its own Kroger data. If you want to know where your items sit against the median and the top decile, and which of your categories are leaking the most shelf, that is the pull.

Summary

• On-shelf availability, not door count, is what turns distribution into sales, and it is the number most launch decks omit.
• Across six brands' Kroger data (about 58 million store-weeks), a distributed item logged an empty shelf in 57% of the item-weeks measured, at 26 empty-shelf events per 100 store-weeks.
• Velocity's spread is wide: the median item does $18 per store per week, the top decile nearly $52, and category matters more than most brands assume.
• Shelf gaps concentrate in heavy, fast-moving categories, and directionally they tend to show up while the DC still holds inventory, which points to the shelf, not upstream supply, as the likely constraint.
• The highest-return move for a brand with strong distribution is usually defending the shelf it already has, not winning more doors.

Related: Kroger banner vs. total · What is ACV? · What is TDP? · Distribution gap calculator