Store operations
How to Compare Retail Store Performance Fairly Across a Network
Comparing retail store performance fairly means ranking stores on what they control, not on where they are. A mall flagship in Gurugram and a franchise outlet in Nashik should not be on the same league table for sales. They can be on the same table for whether the checklist was done, whether the display matched the planogram and whether the advisor presented the offer correctly. The fair method has two halves: adjust the result metrics with cohorts and an index, and compare the execution metrics directly.
Why raw rankings fail
Ranking by sales rewards catchment. Ranking by growth rewards a bad last year. Ranking by conversion rewards low-footfall stores where every walk-in is a serious buyer. Ranking by average ticket rewards the store that carries the premium range. Each of these tables is published somewhere in most networks, and store managers learn within a quarter that the table measures the location, not them, and stop reading it.
A fair comparison has to answer the manager's objection before it is raised: yes, your footfall is a third of the flagship's, and you are being compared with stores like yours.
Build cohorts before you compare
| Cohort factor | Why it matters | Typical buckets |
|---|---|---|
| Format | An EBO, a franchise outlet, an LFR shop-in-shop and a high-street store have different traffic, staffing and control | EBO, franchise, shop-in-shop, high street |
| Footfall tier | Conversion and ticket behave differently at 300 and at 3,000 walk-ins a week | Under 500, 500 to 1,500, over 1,500 per week |
| Store age | A store under a year old is still building its catchment | Under 12 months, 1 to 3 years, over 3 years |
| Carpet area | Range depth and staffing scale with space | Under 800, 800 to 1,500, over 1,500 sq ft |
| City tier | Ticket size, competition and festive timing differ | Metro, tier 1, tier 2 and below |
| Category mix | A store weighted to accessories cannot match a store weighted to hero products on ticket | By share of hero category in sales |
Index against the cohort median
For each result metric, the store index is the store's value over the cohort median, times 100. The Nashik franchise converts at 16 percent against a cohort median of 14 percent: index 114. The Gurugram flagship converts at 18 percent against a metro-mall cohort median of 21 percent: index 86. The flagship sells five times as much and is the weaker performer on the table that matters.
Do the same for average ticket and sales per advisor hour, then combine into one result index with weights that reflect your category: in durables, conversion 50, ticket 30, sales per advisor hour 20 is a reasonable start. Publish the index and the cohort, never the raw rank alone.
Metrics you can compare without adjustment
- Checklist completion on time, with photos.
- Photo audit score and planogram compliance.
- Offer signage accuracy: this week's scheme up, last week's down.
- Capture rate: consented conversations over walk-ins.
- Playbook adherence by step.
- Offer accuracy in the conversation.
- Disclosure rate where the category requires it.
- Number-taken rate in non-buying conversations.
- These are behaviours. A catchment does not change whether the advisor asked for the number, so a network-wide table on these is fair from day one.
The two-column league table
Put the adjusted result index in one column and the unadjusted execution score in the other, and the network sorts into four groups. High result, high execution: leave alone and learn from. High result, low execution: at risk, coasting on catchment. Low result, high execution: the store is doing the job and the problem is range, price or location, not the manager. Low result, low execution: coach, with evidence, this week.
Borentis execution scorecards put BorentisOps checklist and audit scores next to Borentis Floor conversation scores per store; the result column comes from the retailer's POS and footfall data. Without the result column the table says how stores behave; without the execution column it says only where they are.
Publishing the comparison
- Publish cohort rank, not network rank, to store managers.
- Show the evidence line behind an execution score so the number can be argued with fairly.
- Refresh weekly. A monthly table is a verdict; a weekly one is a signal.
- Recognise top movers as well as top stores. A franchise outlet that moved from index 80 to 105 is the story of the month.
Frequently asked questions
What is a like-for-like store comparison?
Comparing a store with stores that share its format, footfall tier, age, size and city tier, so differences in results reflect what the store did rather than where it is. In practice it means cohorts and an index against the cohort median.
How many stores does a cohort need?
At least eight for a median to mean anything. Below that, compare with the closest cohort and say so on the table.
Should franchise stores be compared with company-owned stores?
On execution metrics, yes; the checklist and the playbook are the same. On results, only within the same format, because a franchisee's staffing, hours and local marketing differ from an EBO's.
Can I compare stores across cities?
Yes, with city tier as a cohort factor for result metrics. Execution metrics compare across cities without adjustment.
Related reading
- How to monitor store performance across multiple locations
- How to measure store productivity
- How to identify underperforming retail stores
- Network Performance
Where Borentis applies this
- Execution Scorecards: See the floor before the P&L does.
- Advisor Recognition: Great work, finally witnessed.
- Playbook Adherence: Your playbook, finally observed.
Borentis is the Agentic Operating System for Customer Interactions, built for Indian retail floors: consented one-tap capture on the advisor's phone, every conversation scored against your playbook with the evidence behind every number, leads created when a number is heard, and coaching from your own best conversations.