Retail analytics
Customer Analytics in Retail: Behaviour, Journey and What Customers Actually Say
Customer analytics in retail is the analysis of who a store's customers are, what they buy, how they move through the store and why they do or do not purchase. It is built from three sources: transaction and loyalty data, which describe buyers; footfall and camera data, which describe movement; and surveys, which describe a self-selected few after the visit. What it has rarely included is the customer at the counter, saying what they wanted and why they walked away.
The three questions customer analytics tries to answer
Who are they: segments by value, frequency, category, city, channel. This is the oldest part of the discipline and the best served by data, because every bill and every loyalty scan adds a row.
How do they behave: what goes in the basket together, how often they return, which categories they cross into, how they move between zones. Behaviour analytics reads patterns across many visits.
Why do they buy or not: the question every retailer actually wants answered and the one the data serves worst, because the reason is spoken on the floor and recorded nowhere.
Customer behaviour analytics: what the data can see
Customer behaviour analytics works from records of what customers did. In Indian retail that usually means POS bills tied to a phone number, a loyalty programme where one exists, and increasingly a CRM entry made by the advisor after the visit. The analyses are well established.
- Basket analysis: which products sell together, which drives cross-sell placement and bundle offers.
- Repeat and recency: how many customers return within a window, and how long since the last visit, the basis of RFM segmentation.
- Category migration: whether a customer who bought one category comes back for another.
- Visit timing: which segments come on which days and hours.
- Promotion response: which segments moved on the last scheme and which did not.
Customer journey analytics in a physical store
Customer journey analytics maps the stages a customer passes through and measures how many reach each one. Online, every stage is logged. In a store the stages are arrive, browse, engage with an advisor, decide, and either buy or leave, and most of them are dark.
Arrival is counted by the footfall counter. Browsing can be mapped by cameras or Wi-Fi sensors as zone dwell. Buying is on the POS. The engage and decide stages, where the customer talks to an advisor and makes up their mind, are the centre of the journey and the part no journey map has data for. They are usually filled in with assumptions, a mystery shopper, or a survey answered by the buyers who stayed.
Customer metrics: definitions, sources and owners
| Metric | Definition | Source | Cadence | Owner |
|---|---|---|---|---|
| Repeat rate | Share of customers with a second bill within the window | POS, loyalty | Monthly | Marketing, CRM |
| Segment share | Revenue by segment (value, frequency, category) | POS, loyalty | Monthly | Marketing |
| Basket affinity | Products bought together above chance | POS | Quarterly | Category, merchandising |
| Zone dwell | Time spent per zone per visit | Cameras, Wi-Fi sensors | Weekly | Store operations |
| Engagement rate | Visits that became a conversation with an advisor | Consented conversations against footfall | Weekly | Sales leadership |
| Objection mix | Reasons for not buying, ranked, in the customer's words | Consented conversations | Weekly | Sales leadership, product |
| Unmet request rate | Conversations where the customer asked for something not stocked | Consented conversations | Weekly | Merchandising, buying |
| Non-buyer follow-up rate | Non-buying customers contacted within the window they gave | Conversations joined to CRM | Weekly | Store managers |
Where the conversation adds to customer analytics
The gap in every table above is the non-buyer. In assisted retail, durables, mobiles, jewellery, two-wheelers, cars, most walk-ins leave without a bill, and they also leave without a survey response. Behaviour analytics describes the customers who stayed. The reason the others left is said out loud to the advisor and then lost.
Consented conversation capture closes that gap directly. Borentis records the sales conversation on the advisor's phone with the customer's consent, in Hindi, English or Hinglish, and scores it against the playbook. The by-product for customer analytics is the objection in the customer's words, the competitor they named, the price they were quoted elsewhere and the product they asked for, counted by store and week rather than remembered by an advisor. Joining those conversations to camera events, so a journey map shows what was said at the engage stage, is on the roadmap.
Privacy and consent
Customer analytics on Indian store data sits under the DPDP Act. Transaction data tied to a phone number is personal data, and so is a recorded conversation. The workable pattern is consent before recording, audio deleted after transcription, customers anonymous in every report, and analysis at the level of store and week rather than individual. Done that way, conversation-based customer analytics is on firmer footing than a survey programme that stores names and numbers indefinitely.
Frequently asked questions
What is customer analytics in retail?
Customer analytics in retail is the analysis of who a store's customers are, what and how they buy, how they move through the store and why they purchase or leave, using POS, loyalty, CRM, footfall, camera and survey data, and increasingly consented conversations.
What is the difference between customer behaviour analytics and customer journey analytics?
Behaviour analytics reads patterns across many visits: baskets, repeat rate, category migration. Journey analytics follows the stages of one visit, from arrival through engagement to purchase or exit, and measures how many customers reach each stage.
How do you collect customer data in a physical store?
Phone numbers at billing, loyalty scans, CRM entries by advisors, footfall counters, cameras or Wi-Fi sensors for movement, post-visit surveys, and consented recording of the sales conversation. Each reaches a different slice of customers.
Can customer analytics tell you why customers do not buy?
Not from results data. POS and loyalty only see buyers, and surveys mostly reach buyers. The reason is spoken to the advisor during the visit, which is why consented conversation capture is the source that answers it.
Related reading
- Retail analytics: the full guide
- Footfall analytics
- Retail KPIs only the conversation can measure
- ShopperDNA, on the roadmap
Where Borentis applies this
- Objection Intelligence: The reason they did not buy, in their own words.
- Unmet Demand Signals: Demand for what you did not have.
- Walk-in Recovery: The customer who left is still yours.
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.