Artificial Intelligence technologies are gaining popularity among retailers. Computer Vision helps to analyze visitors’ behavior, store personnel’s performance, checkout lines, availability of goods on the shelves, their popularity, and so on. Among the tasks that the system solves is the management of purchasing behavior using heat maps.
Briefly about the solution
A heat map is a diagram showing the allocation of buyers on the trading floor for a certain period. The map is generated on the basis of video stream analysis, visualizing “popular” and “cold” zones of the store. As a rule, a neural network detects a customer and “remembers” their location in a coordinate matrix. After processing the information, the system uploads a color scheme commonly called a heat map: the brighter the spot on the image is, the more intensive traffic the place has.
What is the data that a heat map visualizes?
Based on a video record of one day, it is possible to build a summary map of visits and divide the indicators by hours – there will be around 8-12 frames. A report based on a video with a duration of 2 days or more can visualize the traffic of visitors for the entire period, as well as show the dynamics of visits to a retail facility by days, hours, and minutes.
The technology makes it possible to distinguish between heat maps of employees’ and customers’ movements and analyze the behavior of both groups. Some analytical services that are used to build a map upload some additional statistics, for example:
- the number of visitors during the day within a given counting interval (3 minutes, 15 minutes, 30 minutes, or more);
Heat maps help to influence buyers’ behavior
So how can we work with the data obtained using a heat map? The most obvious way is to place ads, visitor messages, and promotional items in “hot zones”; or draw up an optimal work schedule for the staff based on the interval map. What is more, heat maps can be used to manage purchasing behavior.
For illustration, let’s consider a couple of strategies.
Placing unpopular goods in the “hot zones”
This strategy helps to increase sales of less popular goods, as well as products that need to be sold in a short time. Thus, it’s not necessary to place a separate rack or draw attention to the shelf by using additional funds that require extra resources.
Relocating popular products to the “cold zones”
When rearranging or replanning the premises is not possible, the solution is to adjust the route from the popular zone of the store to the less crowded. Thus, “colder” departments are “heated” at the least economic cost.
Changing the store navigation in cases stipulated by law
Yes, this refers to the distribution of buyers across the retail space during the spread of an infection. Knowing the location and time of congestion of a large number of visitors, as well as understanding which products are the most popular, one can quickly change the navigation by dividing one “hot zone” into several with the legally required distance from each other.
Optimization of the personnel work schedule, based on statistics that are additionally uploaded to the heat map, organically fits into this scheme.
Changing the location of goods in accordance with the routes of buyers
This strategy works for large stores. By comparing heat maps of different parts of the retail space, it is possible to determine the most popular routes among departments (how the customer flow moves from the entrance to the checkout) and put product categories you wish to promote on particular sections of the route.
Designing trade zones in new stores
When replanning a store or opening a new one, managers can take advantage of the empirical experience of applying heat maps, obtained at existing outlets.
Along with the classic strategies for working with heat maps, the obtained data can be applied from the perspective of the retailer’s individual needs. With a relatively simple and fast implementation of the solution, the business has the opportunity to squeeze the most out of this technology: draw up custom reports, experiment, analyze the experiment results in a short time, easily scale the liked schemes to the entire chain.
E-commerce is not the only direction in the development of retail companies. In some cases, maintaining physical outlets is quite reasonable, and the innovative use of technology will help to bring such sites to maximum profitability.
