Shelf-Gap Economic Priority Score (SGEPS)

GMI-RET-013Retail OperationsPilot-ReadyPattern O-4: Alert Quality / Signal-to-Noise Ratio

Definition

Economic weighting of shelf-gap restocking alerts by revenue at risk per hour, measuring whether out-of-stock response is ordered by financial impact rather than detection sequence.

The Measurement Gap

CV shelf-scanning systems detect out-of-stock conditions and generate restocking alerts in detection order. Store associates work through the queue sequentially or by proximity. Nobody asks whether the order matters economically. It does. A gap on a $2.50 commodity item selling one unit per day costs roughly $0.30 per hour. A gap on a $45 premium SKU selling six units per day costs roughly $33 per hour. If the associate restocks the commodity item first because it was detected first, the premium item sits empty for an extra 20 minutes. Across 50 to 100 shelf gaps per store per day, the mispriorization cost becomes material. Current shelf-scanning analytics report detection counts, average gap duration, and restock completion rates. None measure whether the right gaps were filled first. A store can achieve 95% restock completion and still lose thousands weekly because the sequence was wrong.

Decision Supported:

Should the shelf-gap alert system prioritize restocking tasks by revenue impact, and how much revenue is lost because high-value gaps wait behind low-value ones?

Formula

Primary Formula:

SGEPS = SUM(Revenue_lost_i x Hours_gap_i x Priority_match_i) / SUM(Revenue_lost_i x Hours_gap_i)

Variables:

  • Revenue_lost_i = estimated revenue per hour of out-of-stock for item i (daily_units_sold_i / operating_hours x price_i)
  • Hours_gap_i = time from CV gap detection to restock completion
  • Priority_match_i = 1 if the item was restocked in economic priority order, 0 if a lower-value item was restocked first

Companion Metric (Lost Revenue from Mispriorization):

LRM = SUM(Revenue_lost_i x Excess_wait_i) for items where a lower-value gap was addressed first

  • Excess_wait_i = actual wait time minus estimated wait time if economic priority had been followed

Units: SGEPS is a ratio from 0 to 1, where 1.0 means perfect economic prioritization. LRM is a dollar amount representing the revenue cost of mispriorization.

Edge Cases and Exclusions:

Items with zero historical sales velocity (new items, seasonal reintroductions) should use category average velocity as a proxy. Gaps detected near closing time should be discounted by remaining operating hours. When multiple gaps exist on the same aisle, allow adjacency batching without penalty if the total gap-hours are lower than strict economic sequencing would produce. Require a minimum of 100 shelf-gap alerts per store per measurement period.

Proposed formula, pending validation pilots.

At a Glance

Unit
Score
Trend Direction
Higher; higher-value gaps closed first
Reporting Frequency
Weekly
Category
In-Store AI Operational Effectiveness
← Loss Prevention Intervention Equity Index (LPIEI)Shoplifting Detection Confidence Calibration Index (SDCCI) →

Assess Your Organization

The AI-GPS Assessment evaluates your AI governance maturity and recommends KPIs like this one for your implementation roadmap.

Take the Free AI-GPS Assessment