BrainTrust Query: SKU Rationalization Done Right
Through
a special arrangement, presented here for discussion is a summary of a current
article from Applied Predictive Technologies’ The Insight Aisle blog.
In
a bid to cut expenses and increase revenue, the Girl Scouts are reportedly
testing a plan to reduce their assortments to just six different cookies. It’s
their own version of SKU rationalization.
According to a report in The Wall
Street Journal, "A dozen
councils testing the cutbacks with licensed baker Little Brownie Bakers, which
is owned by Kellogg Co., hope to streamline sales, speed up cookie delivery and,
ultimately, increase profits."
In other words, the Girl Scouts are trying
the changes before implementing them full scale. This is common sense. For
most retailers, correct and profitable SKU rationalization is harder than for
the Girls Scouts. The Girl Scouts’ test
will be focused on one key question: When we reduce assortment, what happens
to the sales of the remaining cookie types, ideally vs. a control group? Reducing
assortment in some (test markets), but not all parts of the country (control
markets) should inform the decision about whether to reduce the cookie assortment.
For retailers, SKU rationalization needs to go a few steps further and it requires
transaction level (T-Log) data, which is not always easy to analyze.
For retailers,
understanding total economic impact goes beyond individual product profitability
measures such as GMROI. It encompasses two criteria about any product on the
shelf that can be learned from the T-Log data and some key questions:
1)Â Product Loyalty:
- How loyal are customers to each product? Have they
demonstrated a willingness to substitute with competing products, or do customers
who buy this product always return for the same SKU? - How does that product loyalty
picture look among the retailer’s
highest value shoppers (most frequent and/or highest spending)?
2)Â Basket Profiles:
- Is this product typically included in higher value baskets?
- Is this product
a "driver" of the trip, or more frequently
an afterthought?
Answering the above questions should inform which products
are candidates for SKU rationalization. As the Girl Scouts show, the final
step is to try the changes in a small, low-risk test before rolling them out
everywhere. Only by testing the changes can retailers measure and understand
customer response and adjust their product assortment accordingly in a full
scale rollout.
Discussion Questions
Discussion Questions: What criteria and steps should guide any SKU rationalization effort for retailers? What are some common mistakes you’ve noticed around recent or ongoing SKU rationalization efforts?
The desire to reduce SKUs and gain a better return for the investment in inventory is certainly easy to understand and a worthwhile goal. As that article points out, it is like many goals it is easy to state, hard to execute.
As the article indicates there is analysis you can do to determine the impact of your SKU rationalization efforts. It also indicates there are issues you want answers to that are not easy to determine. Some of these can be determined over time and through the use as the Girl Scouts are doing–test and control environments.
One of the cautions we issue to clients as they undertake SKU rationalization is that in some case you have to have a decent representation of certain categories to be seen as having the category at all. This is especially true if you are carrying a limited selection in the first place.
Retailers embarking on SKU rationalization often look to their sales and say we will keep just the top X number of items or those that sell only Y units per day not realizing that those other items are why customers to their stores. Customers often want a broad selection even though they may select from the same two or three things.
SKU rationalization is a worthy goal, but like most worthy goals getting there takes work and is more complicated that it may seem.
First, it is unclear how reducing assortment will help the Girl Scouts be more profitable. Before we panic over losing Thin Mints or Trefoils, it will only help them if the baker is going to cut the price to the scouts.
SKU rationalization is an area in which we’ve published a number of papers. We have found that the key is to (a) reduce duplication, as defined by the shopper and (b) consider brand reduction after duplication if there is room to do so. You don’t want to remove products that have a high loyalty index and you don’t want to remove niche products that may keep shoppers coming to that store (so don’t simply look at sales/foot).
SKU rationalization in retail is an interesting dilemma. Before you actually do it, the cost savings from a particular SKU Rat plan are pretty easy to estimate, but the impact on consumers from that precise plan is extremely hard to know. So instead, consultants toss around red herrings like the alleged “paradox of choice,” which is based on weak and not necessarily relevant evidence.
Measuring the actual consumer impact in test is a huge step to get over this problem.
The article above does a good job outlining the fundamentals of SKU reduction. Looking at loyal buyers, not just all buyers; switching, not just sales; and basket, not just item, are key principles.
Girl Scout cookies are in a better position than many retail enterprises that fall into bad SKU proliferation habits:
1. Differentiation. Their cause, salesperson, and heritage already give them a unique position in consumers’ minds. As a result, they don’t need to rely on exclusive items to create differentiation.
2. Product loyalty. Five varieties are 77% of sales. This business doesn’t require them to churn out new varieties each year to stay relevant. Dulche de Leche was not successful in broadening appeal to Spanish speakers.
3. Pricing power. The Scouts have not had to rely on promotions and the resulting demand disruptions to generate sales. In fact, many councils have raised price this year to $4 a box.
They are making the right move to market test before rolling out. But in addition to testing the sales impact, they should verify the cost savings as well. Did they really reduce ingredients or packaging varieties? Did they increase purchasing scale among the components still needed? Did they see inventory holding costs drop?
With thousands and thousands of new items being introduced each year, no retailer can keep adding items forever. While every retailer says they add one item and discontinue one item. This simply does not happen; just ask the DC manager. The norm is to add three items and discontinue two. A few years back there was a study where the retailer discontinued 10% of their items.
The consumers like the increase in items. No retailer can keep adding items for every one lost. Walmart has shown us how not to do SKU rationalization. There is an easy way to do it. What you need is target consumer sales from the frequent shopper program and unit sales from checkout. All showing moving items not purchased by target consumers are candidates for being discontinued.
I commented extensively on this topic in my Views, “The misguided bobbing of the long tail.” I don’t say that “bobbing of the long tail,” AKA SKU-Rat is wrong, just that driving it solely by inventory management costs is seriously misguided. As others have noted, the long tail is ATTRACTIVE, that is, it attracts people to the store/shelf, even if people only buy the most common items, the big head. This matter of attracting is something self-service retailers (merchants, not salesman) have done quite well at. Primarily because their brand suppliers, in the battle for shelf space for themselves, have paid the retailer to accommodate SKU proliferation. So attractiveness has been a consequence, not of skillful selling, but simply of the business model.
The other two HUGE selling issues are navigation and choice, neither of which are done worth squat by self-service retailers (warehouse merchants, not sales people.) The easiest of these to do right is managing the shoppers choice, not by eliminating items, but by TELLING THEM WHICH ONE THEY SHOULD PROBABLY BUY! And that is pretty easy, since you should suggest they buy what most people buy here–it’s in the T-log! But then, it may be easier to just twist supplier arms for a few more big checks than to actually, you know, really learn how to SELL!
Herb Sorensen said it very well–SKU rationalization should not be based on inventory management.
Rather focus must be given to the shopper and her expectations. Too much assortment is most definitely harmful…too little may be even worse.