I recently conducted a poll on LinkedIn asking how best to organize products on a shelf. The results were interesting, with the vast majority of respondents suggesting that the merchandising layout should follow the shopper decision tree. And at one level, that makes sense. But there are some significant issues with this approach. Shopper Decision Trees are an important source of shopper insight, but like most data – it is only a guide. Shopper Decision Trees are NOT the answer. How then should we use Shopper Decision Trees to help us in our merchandising decisions?
There is no single ‘right way’ to layout the products on the shelf
Let’s get this straight from the start. There is no single ‘right way’ to layout products on the shelf. Unfortunately it’s a bit more complicated than that. If anyone suggests that they have an algorithm or Aritificial Intelligence solution that will tell you exactly how to lay out the products on the shelf: I suggest you politely show them the door. Its complicated. It requires decisions. There is no ‘right’ answer. It needs humans, data, experience, knowledge.
Why might shopper decision trees help?
The theory behind Shopper Decision Trees is straightforward. The shopper arrives at the shelf and is faced with lots of products. The shopper needs to find the best product for their need, so we can lay out the products on the shelf to help them do this. For example, what if I’m looking for a refreshing cold beverage to drink on my way home from work. I have lots of needs in there: I want a beverage, a cold one, a refreshing one, a specific brand perhaps? Something healthy or low calorie. And something ‘single serve’ so I can drink it on the way home. So how could we lay out the products to help me find the perfect product?
How shopper decision trees work (in theory!)
Shopper Decision Trees do this by organizing the various needs into a hierarchy: which needs are most important – which needs would be prioritized by a shopper over something else. So perhaps the most important criteria for me is that I want a single serve bottle. And I’d be prepared to sacrifice some of the other needs (for example, low calorie, or a specific brand) to get this. The shelf would then be laid out following all of those principles, grouping all of the single serve products together and separating products from the same brand, separating single serve and multi-serve beverages within the same brand. Separating single serve low calorie products from high calorie variants, etc.
The principles then continue down through the hierarchy of needs. Within the single serve section, the products would be sub-divided into high calorie and low calorie, and then by brand, etc. Each ‘decision’ that the shopper makes is mirrored by the way the products are laid out on the shelf.
The shopper could then work their way down their decision hierarchy: find the single serve products, then find the low calorie ones, then find the refreshing ones within that segment, etc. and so on. Eventually the shopper would find what is the perfect product for their need.
Brilliant eh?
The trouble is, in reality, shopping simply doesn’t work like that! For a number of reasons.
There is no single ‘shopper’ – so there is no single ‘shopper decision tree’
The first problem with this approach is that there is no single ‘shopper’. We are all different. Different shoppers have different missions. Different shoppers have different priorities. You might rate brand above calorie intake. I might be the opposite. So which shopper are we focusing on? Which shoppers Shopper Decision Tree do we follow? Yours or mine?
Of course the lack of shopper homogeneity isn’t new, but it has got worse. Back when I first saw Shopper Decision Trees (twenty years ago) shopping missions were far more homogeneous. Shoppers missions in a particular outlet were, if not the same, more similar. But today we see a wider and wider range of missions in a single store. Any Shopper Decision Tree is going to be ‘wrong’ for some shoppers! An agency output typically tries to focus on ‘the average’ shopper, but of course that means that we’re not optimising for all shoppers.
Most shoppers aren’t deciding – they are finding
The clue to the biggest problem with using Shopper Decision Trees for merchandising lies in the name: Shopper Decision Trees. It implies that the shopper is deciding everything at the shelf. Further, it implies that the shopper is consciously deciding. We know that much of what happens in a shopper’s brain is subconscious. (As an aside – many of the methodologies used to create Shopper Decision Trees are also based on the false assumption that the shopper knows their decision making process – so be careful if you are doing this research!)
But shoppers aren’t purely ‘deciding’. A lot of what they do is ‘finding’. While some shoppers are ‘deciding’ some of the time, most shoppers, most of the time, in most categories, are searching for what they want. Not deciding.
The shopper wants to FIND what they are looking for. A Shopper Decision Tree might show the logical hierarchy of their various needs, but it doesn’t necessarily help a shopper find what they want.
Shoppers navigate stores using lots of different stimulus as they try and cut through the noise and clutter of a store and find what they want. What if the Shopper Decision Tree disrupts these navigational cues?
When Shopper Decision Trees Fail
A few years ago we did some work in the infant and kids milk category. The Shopper Decision Tree suggested that the highest level of the hierarchy is ‘age of child’. That makes logical sense: why would a shopper want to look at products for babies at a different age, with different nutritional needs? So classic Shopper Decision Tree theory would lead to a recommendation to merchandise the shelves by the age of the baby.
But let’s think about the shopper. In this category, the shopper navigates the category by brand. They are quite brand loyal (no parent wants to mess with their babies nutrition without good reason!) the shopper approaches the shelf and looks for their brand. Makes sense right?
But if we have laid out the shelf by age of the child, then the brand might appear in many different locations. Most brands have a variant for each age group after all. So the shopper sees their brand in multiple places. That makes it harder for the shopper to find what they want.
Why? Because the shopper needs a shelf that makes it easy to find products, NOT to decide!
Shopper Decision Trees map the past, not the future
Even if shoppers were deciding everything at the shelf (which they don’t), following the shopper decision tree still wouldn’t necessarily be the best way to merchandise a shelf. As mentioned earlier, the shopper decision tree maps the theoretical hierarchy by which a shopper gets to the best product for their need. So if a shopper decision tree suggested that Brand was at the top of the shopper decision tree then we’d consider merchandising the shelf by brand. But what happens if one of the biggest growth drivers of the category required shoppers to switch brands? Perhaps to get shoppers to a more premium brand? Slavishly following the shopper decision tree would work against this. Merchandising based purely on the shopper decision tree is likely to limit growth by encouraging shoppers to buy what they usually buy.
Shopper Decision Trees are a useful input to shopper marketing and merchandising strategy
Don’t get me wrong: Shopper Decision Trees are a useful input to the merchandising process (and to lots of other shopper marketing activities too). But they are not the answer, and slavishly following the Shopper Decision Tree is not a guaranteed recipe for success. So how best to use Shopper Decision Trees to improve the category layout?
The best practice approach to using shopper research to define how to merchandise a category
Focus on the most important shoppers. All shoppers are different. And some are more valuable than others. Which shoppers are most important? Which are most valuable? Which shoppers would we ideally encourage to change their behavior? Which would ideally continue to behave in the same way? Create clear growth priorities or drivers for the brand or category, and define who the target shopper is for each. Focus on them, not shoppers in general
Use decision trees as an input not an answer. Shopper Decision Trees are not Shopper Decision Trees in the way that many understand them. But they do help us understand the priorities that shoppers have and the trade offs that they might be prepared to make. Don’t throw your Shopper Decision Tree away in disgust: use it as a really useful input to the merchandising/shelf layout process.
Understand the balance between finding and deciding. This is crucial. Shopping is a blend of both finding and deciding, and it varies by shopper, category, mission and channel.
Optimize for growth. Remember that the goal is not JUST to make things easy for shoppers. We are trying to drive growth. That means we need shoppers to change their behavior. Developing a shelf layout is about balance. We want to make it easy for a shopper to do what they want to do, but even easier to do what we want them to do.
The home shelf is arguably the most important part of a retail environment. Displays and signs come and go, but the shelf is there week in, week out. Its hard to get it right, but it is worth the effort. If you want to optimize your performance in store, and create home shelf layouts that deliver real shopper satisfaction as well as growth, check out our training programs or get in touch now to understand how we can help and how we’ve delivered this for other clients.
