Why more insight doesn’t always lead to greater success in-store
On the surface, it sounds like such a simple equation: more insights, better results. After all, the more you know about customers and their shopping habits, the easier it should be to anticipate their needs.
In practice, though, that’s not always the case. Recently, I attended the Category Management Association (CMA)’s annual conference in Dallas. There, I had the pleasure of talking to many retailers and consumer packaged goods (CPG) brands. One of the recurring themes during those conversations was the difficulty that those organisations have when trying to turn insights into actions.
That challenge tends to be particularly acute in the area of assortment. Retailers and brands have a wealth of data to draw from when making their ranging decisions, but often struggle to use that information effectively. As a result, while they might aspire to create a range that resonates and differentiates, the reality is often a little different.
The discussions I had at the CMA event got me thinking, and I believe that there are five main reasons why more insight doesn’t always lead to greater success in-store.
Let’s take a look at each—and what can be done about them.
- Too many assortment tools don’t use the right kind of data
Sales volumes. Inventory days. Margins and profit. All of these datapoints can help retailers and brands understand product performance. The problem is, there’s a real danger that they end up creating self-fulfilling prophecies, too. If “Product X” has been selling well for the past three months, for instance, wouldn’t the smart thing to do be to give it even more shelf space?The issue with using purely commercial data like sales volumes is that it doesn’t provide insight into what customers actually want. A truly effective assortment isn’t just a collection of the best-selling products; it includes a wide variety of niche goods that meet specific needs too. And the only way to understand what those needs are is through deep insight into your customers.Today, the vast majority of assortment tools rely solely on performance data. Our own platform — dunnhumby Assortment — is one of the few to incorporate customer loyalty data. As a result, it’s also one of the only tools that can help you accurately respond to changing customer needs.
- To create success, you need to break down silos
Assortment is a collaborative process, with sales, category management (CatMan), and planogram teams all involved. Unfortunately, those teams are often disconnected—working in silos and making decisions based on very different (and sometimes conflicting) priorities.The sales team will be looking to grab as much space as possible in order to their targets, for instance. CatMan teams will want to ensure that best-selling products are being ranged correctly. And then you have the planogram team, who have the tough job of telling their colleagues that their plans won’t work in the space available.Naturally, that leads to a lot of wasted time and effort—which is why dunnhumby Assortment specifically addresses that disconnect. Assortment provides a collaborative space in which sales, CatMan, and planogram teams can work together.
Moreover, dunnhumby Assortment uses industry-leading artificial intelligence (AI) to create “space aware” recommendations. Machine learning methodologies help to ensure that—as well as being customer-centric and commercially viable—assortments work within the physical constraints of a specific store as well. That leads to greater efficiency and better results.
- Raw data isn’t insight
As noted above, there’s no shortage of assortment-related data for retailers and brands to draw from. Information isn’t the same as insight, though—and it’s the ability to convert raw data into something actionable that makes the difference. That’s why it’s so important to understand the wider market context.Imagine you work for a CPG and data suggests that your category is shrinking by 10%. Is that information useful? Is it actionable? Not particularly.But what if you knew that—while your category was shrinking—your brand was growing? Or, as a retailer, what if you found that one of your categories was growing when it was shrinking at other banners? That is insight, because it gives you a position from which you can act.
Unanchored from anything else, raw data tends to be unhelpful at best—and actively misleading at worst. So, while more insight can drive greater success in store, more data is very unlikely to.
- “Success” is what you make it
Grocery is a fast-moving, highly competitive industry. Because of that, it can be all too tempting to play “follow the leader” when it comes to assortment. After all, if a hyper-localised, store-by-store approach to assortment can deliver results for the best in the business, couldn’t it do the same for you?Not necessarily. Grocery is also a deeply nuanced industry—one in which “success” is contextual. If you operate 50 stores across a concentrated geographic area, for instance, then your assortment priorities are likely to be very different from a banner that sells nationwide. Knowing what problem you’re trying to solve is just as important as solving it. - There’s a big difference between ideal and achievable
Even when you do know what problem you’re trying to solve, how you tackle is dependent on your overall capabilities. Context is important again here—less in terms of what’s happening within the wider market, and more about how your business operates.Say that insight tells you that the fastest route to success is to review and update the category assortment every week at the store level. Unless your business is operationally flawless, that insight is probably useless. It simply can’t be actioned. So, recommendations that are detached from the reality of your business won’t deliver success.
At dunnhumby, we focus on finding the balance between what’s best and what’s achievable.
Would you like to know how dunnhumby Assortment can help you turn data into actionable insights? For more information visit dunnhumby Assortment.
TOPICS
RELATED PRODUCTS
Create customer-centric ranges using AI-powered science
Enable Customer-centric RecommendationsThe latest insights from our experts around the world
![customer first data science analytics & machine learning services](https://www.dunnhumby.com/wp-content/uploads/2023/03/customer-first-data-science-analytics-machine-learning-services.webp)