Thoughts, resources, articles
woman pouring down a brown paint
Photo by on

Analytics is the science of making sense of data to make an informed decision. Retailers have to make sense of the data collected from their various tools to help with the following functions:

  • Demand Forecasting
  • Procurement
  • Logistics & Warehousing
  • Category Management
  • Store layout
  • Shopper Segmentation
  • Pricing
  • Promotions

Retail analytics involves making sense of data to augment decision making processes that impact profitability. The goal of retail analytics can be to increase sales revenues or cut cost (or both).

Level of Investment

Before diving into the types of analytics available, one has to first examine the kind of data available. Unfortunately, data collection is an upfront investment on the retailer’s part.

No/ Bad Investment:

No digital point-of-sale (POS) system or a POS system that doesn’t allow you to extract the raw transactions for analysis work

Basic Investment:

POS System:

  • Transactional Data: Logs sales of products across various time horizons and outlets
  • Product Data: Full record of all products and other related product information (e.g. category, brand, color)

Inventory System:

  • Inventory Data: logs inventory levels of products available across various outlets (or at a central warehouse)

Vendor Management Module:

  • Supplier Data: holds records of products provided by each supplier, the average unit cost and minimum order quantity

Modest Investment:

Loyalty Cards:

  • CRM/ Customer Information: logs and tags shoppers to individual transactions and hold some information about the shopper (e.g. Name, Email)

Large Investment:

In-store Tracker/ Website Analytics:

  • Advanced Tracking Data: logs and tags shoppers how they shop and what they see in the store based on what aisles/ web pages they visited
police blue sky security surveillance
Photo by PhotoMIX Ltd. on

Back to Basics…

Once you have access to these datasets, you will notice that you will always have to start with the basics before you can answer some of the more advanced questions.

An example of more basic analysis work would be a Basket Analysis, “How often is product X also both with Y”. “Given that R is also bought, what is the probability that P is not bought?”

For basic shopper segmentation, you try understanding “When are my best shoppers coming to visit?” or “How much is the average baby boomer spending on weekends?”

More Advanced Stuff…

If you wanted to do more advanced analytics, you would start by combining different dataset. Take the example of the Basket Analysis and Shopper Segmentation from above, a combination would allow you to ask questions like

  • What are working males more likely to buy as compared to teenagers?
  • Are my Gold tier shoppers purchasing the same way as my Silver tier shoppers
  • How are they different?

If you add in tracking data…

  • In which order are these products bought?
  • Should I recommend product Y or X at the point of purchase for shoppers who shop on weekends?

As you can probably tell by now, the ability to combine various datasets allows you to answer questions with much higher granularity. Before you rapidly start searching for new ways to collect data, first decide if you need to that level of granularity for your stage of business and second, whether you have the capacity to react to the findings.

There is no need to track in-store data if you haven’t gotten demand forecasting right and there is no point learning these insights if nobody within the organisation has the capacity to act on them now.

electronics circuit board
Photo by Tookapic on

Final Words

If you have any feedback or questions about this post, we would love to hear from you.

Good luck!