From Dead Zones to Hotspots: Increasing Basket Size by 18% with In-Store Journey Mapping

Our client, a major home goods and electronics retailer, faced a persistent puzzle. Certain high-margin product categories, like premium headphones and smart home devices, were consistently underperforming. Their merchandising team had placed these items in what they believed were prime locations based on conventional retail wisdom—large, well-lit end-caps and dedicated sections. Yet, sales remained flat. They came to us with a fundamental question: do we truly understand how customers move through our stores?

Our team at Iceberg Data saw this not as a merchandising problem, but as a data problem. The client's assumptions were based on gut feelings, not empirical evidence. We proposed a solution to map the invisible pathways of their shoppers using an untapped resource they already owned: their in-store public Wi-Fi network. Our project began by scraping publicly available architectural filings and municipal permit databases to acquire precise floor plans for their two pilot stores. This formed the canvas for our analysis.

The next, most critical step, was data acquisition. With the client's full permission and access, our engineers developed custom scrapers to interface with the web-based management consoles of their commercial Wi-Fi access points. We extracted millions of anonymized connection events over a four-week period. This data was rich, containing a hashed device ID (protecting privacy), connection timestamps, signal strength (RSSI), and the ID of the access point it was connected to. This raw data held the secret to shopper movement.

The real magic happened in the processing. Our data scientists applied a trilateration algorithm to the RSSI data. By measuring a single device's signal strength from three or more different access points simultaneously, we could calculate its approximate coordinates on the store floor plan. By stringing these timestamped coordinates together, we could reconstruct the entire journey of thousands of anonymous shoppers. Aggregating these journeys allowed us to generate a vivid, data-driven heatmap of each store, as referenced in our `Example_Output_JSON` .

The results were a revelation for the client. The heatmaps immediately falsified their long-held beliefs. A key finding was that the premium headphone display, located at the back of an aisle, was one of the store's most significant `cold_zones` . The data showed a clear pattern: customers would walk down the main thoroughfare, get drawn into the brightly lit video game section, and then turn back, completely bypassing the adjacent aisle where the headphones were located. Our report identified this `aisle_7b_headphones` zone as having an average dwell time of just 12 seconds.

Conversely, we discovered an unexpected `hot_zone` . The wide, transitional space leading towards the checkout queues, an area the client had used for little more than holding shopping carts, showed incredibly high traffic and long dwell times (`avg_dwell_time_sec` of 110). Shoppers milled around here, waiting for family or simply composing themselves before the final purchase. It was a captive audience they were completely ignoring.

Armed with our `heatmap_data`, the client executed a bold, data-driven store redesign. They moved the entire premium headphone category from the 'cold' aisle to a new, sleek kiosk built in the 'hot' pre-checkout zone. The former dead zone was repurposed for low-consideration clearance items. The impact was staggering. In the two pilot stores, sales of premium headphones shot up by over 40% within the first two months. More importantly, by placing high-margin, impulse-friendly products directly in the path of a captive audience, the store's overall average basket size increased by 18%. Our analysis didn't just tweak a display; it fundamentally reshaped their understanding of in-store customer behavior, turning underutilized floor space into a significant revenue driver.