Blog · 23 September 2026
Out-of-stock dashboard: which indicators to track every week?
In shortAn out-of-stock dashboard fits in five weekly indicators: the on-shelf out-of-stock rate per aisle (counted on the shelf, not in the system), the average duration of an out-of-stock, the ten most often missing products, estimated lost sales, and the share of out-of-stocks for which book stock was positive (phantom stocks). Each points to a different lever: struggling aisle, replenishment speed, facing to review, economic priority, stock reliability.
Why five indicators and not one
An overall out-of-stock rate does not say what to do. It hides the aisle that is doing badly, does not distinguish a one-hour out-of-stock from a one-day one, and does not say whether the cause is upstream or in the store. Five indicators, counted every week, are enough to decide.
1. The on-shelf out-of-stock rate per aisle
Calculation: absent references ÷ expected references × 100, counted on the shelf at a fixed time, twice a day, then averaged over the week per aisle. The method is in on-shelf out-of-stock rate: how to calculate it.
Reading: compare each aisle with its own previous week. An aisle going from 5% to 8% has a new problem (delivery, absence, promotion); an aisle at 9% for a month has a structural problem (facing, replenishment). Benchmark: French grocery stores sit between 5 and 8% according to the ECR France / IRI barometer, and ECR UK’s measure found about 92% on-shelf availability on UK fast sellers; under 3%, an aisle is very well run.
2. The average duration of an out-of-stock
Calculation: time between noticing the gap and refilling it, averaged over the week. Without continuous detection, approximate it with the two daily counts (absent at 11 am and at 5 pm = at least six hours).
Reading: duration measures reaction speed, not ordering. An average duration above three hours points to replenishment concentrated in the morning; the organisation of passes is described in how to organise replenishment to avoid gaps on the shelf. It is the indicator that moves fastest when you act.
3. The ten most often missing products
Calculation: number of out-of-stock observations per reference over the week, ranked.
Reading: these ten references concentrate a disproportionate share of lost sales. A product in the top 10 three weeks running has a structural cause: facing too short, reorder point too low, or phantom stock. The list feeds directly the facing review (facing and shelf capacity by sales velocity) and cycle counting.
4. Estimated lost sales
Calculation: for each out-of-stock, hourly sales velocity × duration × 55% (share really lost according to the shopper reactions measured by Gruen, Corsten and Bharadwaj), then summed over the week. The method is detailed in how to measure sales lost to out-of-stocks.
Reading: this is the indicator that speaks to the accounts and arbitrates priorities: an out-of-stock on milk costs more than one on a slow mover, even though both count as one in the out-of-stock rate.
5. The share of out-of-stocks with positive book stock
Calculation: out-of-stocks observed while the reference’s book stock was above zero ÷ out-of-stocks observed.
Reading: this is the share of out-of-stocks the inventory system does not see, hence that automatic ordering will not fix. It measures book stock reliability and the weight of the back room. Above 30%, cycle counting and back-room organisation are the first levers (cycle counting: twenty minutes a day).
The one-page dashboard
| Indicator | This week | Previous week | Lever if worse |
|---|---|---|---|
| Out-of-stock rate per aisle | per aisle | per aisle | aisle to treat |
| Average duration of an out-of-stock | hours | hours | replenishment passes |
| Top 10 missing products | list | list | facing, reorder point |
| Estimated lost sales | euros | euros | priorities |
| Share of out-of-stocks with positive book stock | % | % | cycle counting, back room |
Five lines, one page, fifteen minutes of reading at the Monday briefing. The rest of the time goes on the levers, not on the figures.
What shopper reporting changes
The five indicators need one thing: knowing which reference was missing, when, and for how long. Shopper reports with ShelfAlert provide that data continuously, without a counting walk, and the dashboard calculates them for each store, aisle by aisle. 14-day trial, no payment card: see the plan for store managers.
Sources
- ECR France / IRI out-of-stock barometer: LSA, out-of-stock rate.
- ECR UK / IGD on-shelf availability measure: On-shelf availability: the case of a UK grocery retailer.
- Gruen, Corsten, Bharadwaj, “Retail Out-of-Stocks: A Worldwide Examination”, GMA, 2002: full study (PDF).
- DeHoratius, Raman, “Inventory Record Inaccuracy: An Empirical Analysis”, Management Science, 2008: abstract.