Asking simple questions of your scan data can help improve operations and increase sales.
By Jeremie Myhren
Most convenience stores today are collecting scan data, but that’s only half the battle. Knowing how to maximize your data analysis to drive sales, best the competition and improve your bottom line is the ultimate goal.
In order to continue down the path to that goal, start thinking of what questions you wish to ask of your data, and ensure it can answer them for you.
Today, many convenience stores are using scan data to complete a thorough consumer analysis, which can bring strategic advantages.
Tracking your consumers’ shopping habits allows you to target them with offers that better resonate with their preferences, which means it’s more likely to inspire sales. Evaluating your competition’s strengths and weaknesses and then measuring that against your own can likewise reveal new opportunities within your own business plan.
ASKING QUESTIONS
To further improve your bottom line, ask the following questions of your company’s scan data:
• What are we going out of stock on? What is the margin impact of those lost sales? Using basket analysis, what items that normally pair with the out-of-stock items are also not being sold as a result? Communicate these findings and understandings to your teams and suppliers.
• Are we suffering from phantom out-of-stocks? Merchandise stuck in the backroom or in wholesaler’s tote is as bad as an out-of-stock, if it is not on the shelf ready to sell. Reporting on items that commonly sell and show as in-stock but have gaps in sales can help you identify these costly phantoms, and as a result you can better optimize key delivery times and your labor on hand to unpack those deliveries.
• What items have a statistically higher than average association with each other? Knowing the common associations, you can create relevant combination offers and ensure products are placed in a way that removes friction from a customer’s ability to buy them together. Don’t make the mistake of only looking at this globally. Look to individual stores to understand what works. In some cases, you’ll discover associations in a single store that are a result of accidental optimization that you can then apply to the rest of your chain.
• What patterns exist that you have not recognized or could not possibly recognize on your own? Big data and machine learning related developments have created new ways of understanding and responding to data. Perhaps four of your stores are within a one-mile radius of a school. You may not even be aware of those stores shared association with schools, but that association is hiding in your scan data. Machine learning algorithms can detect those patterns, helping you to then merchandise and market locally based on them.
• Are certain key products/categories lagging in certain stores as compared to your chain overall, or as compared with your best performing stores? Maybe it is time to invest in some new ice chests, or change the way high value items are merchandised in certain areas to encourage sales and reduce friction for the customer who may be a potential buyer.
• What items in a given store haven’t moved in the last quarter? How about the last year? How about longer? Even the most sophisticated retailers who do a great job of working with their scan data sometimes overlook this simple way of looking at scan data and have been surprised to see items on-hand that haven’t moved in longer than five years. How much of your working capital is tied up in dead inventory that makes for full shelves but is otherwise doing nothing for you?
