We now know – or better, have all the means available to know – that less could be much more, in terms of choice. (see -> the books about customer’s choice and a famous “marmalade” experiment).
But back then in the pre-millennium times, Slovenian retail was still in the transitional stage, on the way out of the socialist model and some things were not so obvious.
Those were the early days of computer-aided planogramming. Not only in the region but also worldwide. As a freshman in the professional world, I unexpectedly encountered a task to arrange, ehm – declutter! – the coolers for the number one national petrol gas chain operator named suitably Petrol. There were all sorts of additional conditions set both from my task-giver and also employer Coca-Cola and the obligations towards the key account Petrol.
Fortunately, I didn’t know much about the process – never even heard about a thing called space management during the university courses – so I approached it pragmatically and learned one-the-go while listening attentively to each hint from some older, refined Coca-Cola cats who really didn’t know much about computers but knew a whole lot about sales.
- Went down to the station, talked in the back-office with the owner for a while, sipped coffee, then studied the coolers, measured them and took the pictures of the initial situation.
- Then I loaded one product of each drink into a cardboard box and took it back to the office. I measured it, height, width, weight, wrote down all sorts of parameters like flavor, category, package, unit case etc. This was the foundation on which I built a database.
- Took the boxes with the products back to service.
So far so good. Only that when I checked the coolers at the gas station nearest to the offices, its contents were completely different than the first time. In the meantime, some new products were added, some previously encountered now missing. Later I’ve found out that the latter occurrences have a name – out-of-stocks.
Not knowing exactly how to approach these daily variations, I moved over to collecting and analyzing sales data for the whole chain of test stores (10 out of 300). The absurdity increased. The data showed that there were more than 500 different articles on-stock while the average cooler was 180 centimeters long. Something didn’t fit, right? Limited space, too many products. Plus: it was physically impossible to squeeze all the products into the cooler with 4 levels of shelves.
With software – old stand-alone, dongle licensed, Intactix (later JDA) – I’ve made an analysis.
- The first 10% of products made almost 90% of total sales
- More than half of the open products together didn’t make up for 10% of revenue
It would be only later when I got used to such sales distributions pretty common in all retail categories, especially those frequently visited.
The boss examined the figures and nodded. Close to what he expected. OK, a little bit worse. I can prepare a new proposal. Just cut the superfluous assortment down – of course, as long there is not “our product” on the list. So I focused = limited assortment.
- Cut the loose tail of the products
- Found out how important distribution (=how many stores actually sell a product) and the date of the first introduction are.
- Also found out that besides the sales figures sometimes a little “something else” matters and that “something else” is in custody of the retailer. It took me a few more years to solve that mysterious “something else” puzzle – but that’s not the case here.
- I have cut down heavily – but not nearly as much as it would be later shown to be even closer to optimum.
- Introduced the list to a) bosses, b) client’s representatives.
- Made corrections. Included some of the products that should be out.
- Prepared a programme to clean up, stop ordering and selling the excluded products.
- Introduced it to the field managers which were eventually responsible for communication with store managers.
There were resistances on different levels. We cut down the assortment through stages. It took us one year to get from 500 to 220 regular SKU’s, then an additional 6 months to get it down to 150. Finally, we achieved 120 SKU’s.
The sales? Growth indexes not only didn’t fall. Went up. Easier navigation, better brand exposure, fewer out-of-stocks and – perhaps most of all – simplified choice brought to a 140 index. Ok, GDP grew as well and the number of visitors should be taken into account. But overall figures were much better than the market. Less is more. When you have 500 different products which the shoppers don’t really see as 500 options to evaluate – but more of at least 400 reasons not to find (or even buy) the product you really seek.
