OSA, share of shelf, planogram, price, POSM, Perfect Store score: the six commercial KPIs you can measure from a single shelf photo, with benchmark ranges.
Commercial KPIs in retail execution are the measurable links between your sales strategy and the physical shelf. Targets are set in headquarters; results are produced in stores. The KPI layer is what tells you, store by store and week by week, whether execution is supporting the plan, before the quarter closes and the question becomes rhetorical.
This pillar guide covers the six KPIs that matter most in shelf execution, how each connects to revenue growth, the benchmark ranges to expect, why the measurement method decides whether the numbers are trustworthy, and how to build the operating rhythm that turns a dashboard into decisions. It is written for commercial directors, trade marketing leaders, and category managers who own a number, and need the shelf to deliver it.
Why most KPI programs fail before they start
Three design flaws account for most disappointing KPI initiatives. First, too many metrics: a 23-KPI scorecard guarantees that none of them owns anyone's attention. Second, unmeasurable metrics: KPIs defined in a workshop but collected manually arrive late, partial, and disputed, so meetings debate the data instead of the decision. Third, disconnected metrics: KPIs that don't ladder to a P&L line get deprioritized the first time budgets tighten.
The fix is the same in all three cases: a small set of automatically measured, financially bridged KPIs. That is exactly what shelf-photo recognition makes possible, and why the six below have become the standard core.
The six KPIs one photo can measure
|
1. On-Shelf Availability (OSA). Are the planned products physically present, in the right variant and quantity? The most direct sales link in the set: roughly every 3% of OSA equals 1% of sales. Programs typically gain +5 to 15 percentage points of availability once OSA is continuously measured; real-time detection runs at ~96% accuracy with absence reasons logged per SKU. |
2. Share of Shelf (SoS). What fraction of the category's space, in facings or centimeters, is yours? Space share tracks market share in most categories. Continuous measurement typically yields +3–7 pp within six months, with category boundaries detected automatically at SKU, brand, and category level. |
|
3. Planogram compliance. Does the executed layout match the agreed plan? Worth +1–2% of sales from optimal layout, up to +7% in placement-sensitive categories, measured at 98%+ category-level data accuracy with violations classified by error type. |
4. Price execution. Are regular and promotional prices displayed correctly, inside the agreed bands? Recognition reads tags at up to 90–97% accuracy and cuts monitoring cost ~60% versus manual collection, with 100% on-shelf pricing accuracy as the operational target. |
|
5. POSM presence. Are the materials you paid for installed where you paid for them to be? The KPI that turns trade-spend ROI from an assumption into a measurement. |
6. Perfect Store score. The weighted composite of the five above, one number per store visit that ranks the network and prioritizes the field's time. |
The composite deserves one more word. The Perfect Store score is not a sixth measurement, it is a prioritization device built from the other five. Its job is to rank the network so that the scarcest resource in trade, field time, flows to the stores and gaps where it pays most. Read individually, the five KPIs diagnose; read as a composite, they triage.
Why measurement method decides everything
Manual KPI collection forces a permanent trade-off between coverage and accuracy: more stores means more hours means more transcription errors. Image recognition removes the trade-off. Every visit produces all six KPIs from the same photo, with proof attached, in under 20 seconds, and total visit time drops 40–60% because the audit stops being a separate activity. The audit becomes a by-product of the visit.
The reliability numbers matter because KPIs are used to manage people and negotiate with partners: ~97% product recognition accuracy, automatic photo quality control, GPS-verified visits. A KPI nobody can dispute changes meetings, the conversation moves from "is this number right?" to "what do we do about it?"
|
1 One photo in the Ailet app calculates all six commercial KPIs simultaneously, availability with absence reasons, share of shelf in facings or centimeters, planogram compliance by error type, price execution including promo mechanics, POSM presence, and the composite Perfect Store score, in under 20 seconds, online or offline. |
2 |
3 The end-to-end cycle from shot to insight runs in under 20 seconds. |
From lagging to leading: using shelf KPIs as a forecast input
Because the six KPIs move before sell-out does, they are forecast inputs, not just scorecards. A sustained OSA improvement in a region reliably precedes a volume lift there; a share-of-shelf erosion in a key chain precedes a share loss; a price-execution breakdown during a promotion predicts the promotion's miss before the sell-out data confirms it. Commercial teams that feed shelf KPIs into their S&OP and demand-planning conversations consistently report two effects: forecasts argue less (the leading indicator settles the debate early), and corrective action starts weeks sooner, which is often the entire difference between saving a quarter and explaining one.
A practical starter: add one slide to the monthly demand review showing the three-month trend of OSA and SoS against actual sell-out per region. Within two cycles the correlation in your own data becomes visible, and from then on the shelf KPIs carry forecasting authority no consultant's benchmark could give them.
The competitive dimension: the same photo reads the whole category
An underused property of photo-based measurement: recognition does not stop at your products. The same image that scores your availability and facings also reads the competitive set, their share of shelf, their pricing, their promo presence, their POSM. That converts the KPI layer into a lightweight competitive-intelligence system with zero incremental field cost: category share-of-shelf trends by chain, competitor promo intensity by region, pricing position store by store. Commercial planning with that input negotiates differently, you arrive knowing not only your own execution but the category's.
Bridging KPIs to the P&L: the table your CFO wants
Each KPI earns its place by mapping to a financial line. The bridges that survive finance review:
|
OSA → recovered sales (OSA points gained ÷ 3) ≈ % sales recovered. A six-point gain on a 50M category ≈ 1M. |
SoS → share defense Facings recovered × revenue per facing, the cheapest share growth in the portfolio because the distribution already exists. |
|
Planogram → layout uplift +1–2% baseline, up to +7% where placement sensitivity is proven by your own compliance-vs-sell-out data. |
Price → protected margin Error-weeks eliminated × average margin impact, plus promotions verified live. |
|
POSM → trade-spend ROI Spend on verified-installed materials vs spend on missing ones, usually the most uncomfortable and most useful number in the first quarter. |
Audit efficiency → cost line –60% audit cost; audits in ~2 minutes per category; field hours redeployed to selling. |
From dashboard to decisions: the operating rhythm
KPIs change outcomes only when they are wired into a cadence with owners:
- Daily, field: each rep opens the visit with the store's KPI gaps as the task list. No generic checklists; the store's own data drives the work.
- Weekly, management: exception review of the bottom quintile per KPI and per region. The recurring failing KPI per cluster is the coaching agenda.
- Monthly, commercial leadership: KPI trends vs sell-out by chain; the negotiation pack per key account (photo evidence of agreed-vs-actual); weight rebalancing if the market shifted.
- Quarterly, strategy: the KPI-to-P&L bridge updated with actuals; targets reset; the category playbook revised with what the data proved.
Channel weighting: one scorecard, different math
A convenience store and a hypermarket should not share KPI weights. Availability and price dominate small formats; placement and POSM carry more weight in large ones; the traditional channel emphasizes presence, equipment, and visit verification. Run one architecture with channel-specific weights and report side by side, the comparison itself is informative, and it keeps the field's targets fair, which keeps the data reliable.
For European teams: KPIs as strategy, not inspection
The framing that earns executive sponsorship connects each KPI to business performance rather than field discipline: OSA as recovered revenue, price execution as protected margin, POSM as trade-spend ROI, the Perfect Store score as the operational expression of category strategy. Data-driven, strategic framing is also what your buyers respond to, retail KPI tracking presented as a margin and growth lever, not a compliance hammer.
|
1. Sequence matters Launch availability first, prove the recovered-sales bridge, then add the remaining KPIs. Each KPI inherits the trust the previous one earned, building credibility faster than big-bang launches. |
2. Visibility breeds usage Put the KPI dashboard on the sales floor's shared screens, not just in management reviews. Measurement that is public, fair, and photo-backed becomes culture. |
Common mistakes
|
Scorecard inflation. Six KPIs with owners beat twenty without. Add metrics only when a decision needs them. |
Averages without distributions. Network averages hide the bottom quintile, which is where the money is. |
|
KPIs without tasks. Every gap must generate an owned action in the field app or the account plan, otherwise the dashboard is scenery. |
Annual targets on weekly metrics. Shelf KPIs move weekly; targets and reviews should match the metabolism. |
|
Trusting self-reported data. If the KPI feeds a bonus or a negotiation, it needs photo verification behind it. |
FAQ
- Which KPI should we start with?
- OSA, it has the most direct sales link (≈3% OSA = 1% sales), the fastest payback, and the simplest story for leadership. Add the others once the availability loop is running.
- How often should commercial KPIs be reviewed?
- Field level daily (in-app), management weekly (exceptions), executive monthly (trends and bridges). Quarterly reviews alone are too slow for shelf metrics.
- How do trade marketing KPIs differ from sales KPIs?
- Sales KPIs measure outcomes (volume, value, share); trade marketing KPIs measure the in-store conditions that produce them. The six in this guide are leading indicators, they move weeks before sell-out does.
- Can all six really come from one photo?
- Yes, recognition identifies SKUs, positions, facings, prices, and materials simultaneously, so each KPI is a different calculation on the same recognized image, which is also why adding KPIs adds no field time.
- What does the data architecture need?
- Photo quality control, GPS-verified visits, planogram and price-list version control, and BI dashboards with drill-down to the photo. Auditability is what makes the numbers negotiation-grade.
- How do we keep the field from gaming the metrics?
- Photo verification plus automatic quality control removes the main vectors; tying part of the bonus to improvement rather than absolute level keeps low-base stores motivated.
- How does this connect to ABM and sales outreach?
- Engaged accounts respond to KPI framing more than feature framing, benchmark data per KPI (the ranges in this guide) gives sales teams a consultative opener that positions the platform as a revenue instrument rather than an audit tool.
- Engaged accounts respond to KPI framing more than feature framing, benchmark data per KPI (the ranges in this guide) gives sales teams a consultative opener that positions the platform as a revenue instrument rather than an audit tool.
