One line take-out: Each of us has a very different opinion on what the store should stock. To win us all stores need a wide range.
The top selling 1000 items in a supermarket generate about half of its sales revenue. Which means that it’s vital that store managers make these items easy to see and buy – but that’s another story.
What I’d like to highlight today is that the other 30,000 or so items they stock sell very little volume. This is what is sometimes called “the long tail”.
Stores try hard to weed out items that don’t sell. So the typical store item does sell, but rarely. Stores are full of stock that barely moves while a tiny percentage of the items fly off the shelf.
This can lead marketing consultants to advise retailers to pare back their range to concentrate on the items that deliver most of their revenue and profits. Yet this range (and cost) cutting strategy often fails. Unfortunately, it’s been encouraged by recent research (some of it flawed) on consumer confusion – research that mistakenly suggested that smaller ranges will increase sales.
It’s true that stores look cluttered and complicated. The average household only buys a few hundred different items from a supermarket in a year. That is, they do a lot of repeat buying of some items over and over. So each buyer is looking for a few things out of the 30-50,000 on offer in the store. That makes shopping sound like a horribly complicated task.
So why on earth would consumers be attracted to stores that stock so many items – most of which they don’t buy? One notion is that consumers like the IDEA of choice, that they are attracted to variety but once they actually arrive in-store they fall back on their habitual nature and existing loyalties.
There may be a little truth in this explanation but the real reason is that consumers are very heterogeneous in the items they buy. Remember that all those items in the store do sell, each item has its buyers. So given that each of us is buying only a tiny proportion of the items in store the odds that my shopping basket will share anything in common with the person in front of me in the queue (or anyone else for that matter) is very low. As I often point out, if you look at what’s in the shopping trolleys of fellow shoppers you see that “other people buy weird stuff”, or at least that they buy different items from you.
The few items in common in any two trolleys are, of course, most likely to be those items that sell in large volumes. These will appear in many more people’s trolleys. Even so most of the items in our trolley will not be from the ‘top 1000’ and so hardly anyone else will buy them.
The Double Jeopardy Law tells us that an item with low market share will be repeat-bought less often than its rivals, but not dramatically less often, the main reason that it sells so little is that few people ever buy it. Which means that many of the many low selling items in a supermarket are, in effect, being stocked for just a few consumers. Some may even be stocked for a single household. But for these few buyers these items are important, they buy them, maybe not that often (but that’s true of most things we buy), they know them, they are in their heads and their pantries – but not many other people’s.
Because we buy these items we like stores that stock them. We each enter a store looking for “our stuff”. If the store doesn’t stock the things we buy we can sometimes find ourselves inconvenienced. We want to see, and be able to find, the items of interest to us. That makes a store attractive to us. Fortunately for store managers consumers are extraordinarily good at filtering out all the brands and SKUs that aren’t in their personal repertoire and finding their brands. Successful stores make this even easier for consumers.
So my point is don’t make the mistake of thinking that a store can do without 90%+ of its range. Stores compete for shoppers, and shoppers vary enormously in what they look for, in what mental structures are in their head, in what they see. Each of us has a very different opinion on what the store should stock. To win us all stores need a wide range.
Marketing Science 
This theory has two soft points:
a) it assumes that small items are largely bought on purpose and are without alternative. In reality there is hardly anything for anybody that is without alternative and therefore reducing item X will likely lead to buying item Y, not to consumers leaving the store. The art of stocking is about identifying those items out of the “long tale” that have a low likelihood of people finding an alternative. This does not mean that SKU’s can be reduced by 90%, but 10-20% could already be big help.
b) there is a cost coming with every SKU behind complexity, handling and lost opportunity for this space. The decision on which SKU’s to carry should look at return, not only at potential incremental sales, return obviously taking likelihood of leaving a store and total basket into account. This bar is higher than just looking at chance of selling.
I didn’t write “leaving the store” did I? I hope I wrote that finding the items that are in your repertoire makes the store more attractive. More likely to be chosen.
If my store took away any of the items I buy I would (at first) simply buy another similar item. But, it’s a slippery slope for the store. Now other stores have become that little bit more attractive relatively. Reducing range doesn’t guarantee I can find what I want faster, it can even slow me down. The harder it is to find my stuff, the slower I shop, the less I spend, the more likely I am to go elsewhere.
I think there is a parallel in the cable and sat. TV business in the US: most subsribers watch relatively few channels(average under 20 out of 200+). But most of the channels that are rarely watched overall, are very important to some subscriber households, and, maybe more importantly, to one person in a larger number of subscriber households.
Thanks for a great post. I’m in the grocery business myself and faced with the very problem you’re referring to. The question I have is: Where do you cut the long tail? Because it needs to be cut somewhere, doesn’t it? Or should you let it go on…forever?
Very good question. Add (or don’t remove) any product from the tail so long as it wins some customers. It doesn’t have to sell much to still be important to some customers – who of course buy lots of other stuff.
So the key to deletions are products thst people are more than happy to always substitute something else, ideally without even noticing. Typically the non-brands (often non advertised).