Sightcorp logo

Retail Analytics

Everything about Retail Analytics


Retail Analytics

Retail analytics have been available for some time now in brick and mortar stores. However, the spotlight for data gathering and analysis has been in the digital world.

Retail today is being shaped by the empowered customer, who demands convenience, customization, collaboration, and consistency. To deliver this demand to the customer, retailers employ retail analytics, which offers insight into how well marketing is working, what customers actually do when they enter a store and whether there is an end result to their visit, such as a sale (conversion).

What are Retail Analytics?

Retail analytics offer insights into factors such as traffic counting and retail conversion rates that provide tangible proof of a store’s successes or failures. Essentially, retail analytics is used to help make better choices, run businesses more efficiently, and deliver improved customer service analytics. Essentially, companies use analytics to get a better picture of their target demographic.

What are the most commonly deployed analytics solutions in retail?

Before the recent advancements in retail analytics, marketers’ ability to track the performance of media and promotions in driving in-store traffic, sales and brand recognition were largely limited to analyzing sales and traffic trends. Now, however, there are a number of ways that analytics can bring powerful insights to a retail organization. Here are the most commonly used analytics in retail:

People counting:
People counting analytics are essential to a retailer as they give information about the number of visitors to a store at any given time or during some time period. From this data, a retailer is able to extract and understand information about how many people entered the store and how many left without buying. When combined with sales data, retailers can then calculate the store conversion rate.

Hot zone and dwell time:
Knowing where people go and stay within a store is also of extreme importance to a retailer. This is known as hot zone and dwell time. This trajectory analysis can enable store managers to optimize store layout for better product placement. In addition, this data can be used to evaluate or enhance the effectiveness of sale or advertising displays.

Customer behavior:
Customer behavior analysis is less statistical in nature, but it can provide finer-grained information about how customers interact with products and advertisement in a store. This information also leads to a better understanding of the customer’s buying decision.

Gaze analysis:
Gaze analysis is vital to understanding customer behavior in terms of attention, feeling and desire. When a person is in the field of view of a static camera, inside a digital Point-of-Sale screen for instance, gaze can give information about the focus point of that person. From here, retailers can learn about what is attractive about the advertisement and how long it was looked at. This can also be applied in store shelves. Consumer attention estimation can help retailers define the optimal position of products in shelves and collect statistics about the most interesting products and several other applications.

Understanding the demographic makeup of shoppers at stores across a chain or at particular locations assists in determining the success of campaigns designed to engage target segments. This means that advertising messages displayed through digital signage, placed at Point-of-Sale for instance, can be tailored to a specific target audience, which increases engagement because of the relevancy of the message.

Discover Our Products

You might also find the following articles interesting:



Technical Specifications

The table below shows how FaceMatch SDK performs on the Labelled Faces in the Wild (LFW) dataset:

FPRTPRThreshold (Inverse of distance)
0.10.99900 ±0.002130.55448
0.010.99667 ±0.005370.59791
0.0010.99367 ±0.006050.62989

FPR = False Positive Rate
TPR = True Positive Rate

These results are an indication only and are based on the specific dataset Labelled Faces in the Wild. Customers can expect similar performance, with possible variations due to hardware and the availability of annotated data.