Getting SKU rationalization right
Through a special arrangement, presented here for discussion is a summary of a current article from Frozen & Refrigerated Buyer magazine. A long-time Harris Teeter executive, Mr. Harris is a former chairman of the National Frozen & Refrigerated Foods Association and a member of the Refrigerated Foods Hall of Fame.
SKU rationalization can make or break your department but there are ways to do it while limiting the disappointment of scratched items to many of your customers.
Here are some tips to help you get it right:
1. Set a benchmark. When I was a category manager, I had a rule of thumb that said if a SKU didn’t sell a half a case a week per store, it was subject to being cut. If I got too many complaints, I’d put it back.
2. Don’t be afraid to make a decision out of fear you’ll be second-guessed from above. You’ve got to keep trying new programs, new ideas. If the boss gets on you too hard, all you have to do is say "I’ll fix it."
3. Go with your gut. Special item requests by customers are not easy to execute. I’ve always believed that if an item is slow for you, it’s probably also slower for your competition and even the manufacturer. It really should be up to the manufacturer to get rid of it — it’ll save a lot of grief for everyone.
4. Go easy on carrying multiple sizes of everything. Do you really need three sizes of parmigiana cheese or Velveeta? You may decide to get rid of the smallest size and let those shoppers go to 7-Eleven. They aren’t likely to buy a lot of groceries anyway, and you really have to get rid of duplication.
5. Accept new items when they come out without waiting six months for a formal category review. If you review a category once or twice a year, that’s fine. But don’t shut down the possibility of profiting from something new just because you want to wait for your next category review.
6. Don’t lock into two brands and a private label per category. If I have to have 15 brands of yogurt to satisfy my customers and keep them loyal, so be it. Do I carry every item they have? No. Do I carry half of them? Maybe. Don’t send your customers away.
7. Bring in new products based solely on their merits and nothing more. Lots of times, if a category manager doesn’t like a particular manufacturer or broker rep, the item just doesn’t get authorized. That happens a lot. This can really hurt your business. Have I ever done that? Yes. Was I wrong? Definitely. Eventually, I had to add the product anyway.
Discussion Questions
What are the most common mistakes made when it comes to SKU rationalization at retail? Would you add any tips to effectively managing SKUs to those offered in the article?
These are all good tips. However, I find that generally speaking, the normal approach to SKU rationalization tries to be so rational that it becomes irrational. When retailers follow the usual SKU rationalization guidance the result is that all retailers look exactly the same with very few shelf points of differentiation. In the end, the only differentiation between Retailers “A” to Retailer “B” is price.
Retailers need to carry some of the niche items, specialty items, premium items, new items, and innovations that not everyone else necessarily carries, even if the benchmarks in sales velocity are not met. Why? Because retailers, you need to draw some destination traffic of the most prized consumers that come to the store to buy that special item that she can’t necessarily find across the street, or at the convenience store, or at every other supermarket and big box store. While she is in your store she is likely to fill her shopping basket with her normal items plus she will buy some of the highest margin items in your store and you just might make a regular customer out of her.
Being afraid to cut deep enough, followed by cutting too deeply.
One of the most common mistakes in SKU rationalizations has to be “shooting the short guys.” In other words, ranking all the items in the section/category and delisting the slowest sellers, regardless of what they are. Often the third SKU of chocolate will outsell the first SKU of blueberry — but that blueberry SKU might be 100% incremental to the line while the third chocolate SKU is only 20% incremental. Shooting blueberry can cost you more sales and disappoint consumers.
I would think his first point, set a benchmark, is a common mistake. That says you don’t understand what your customers want and you’re banking on them telling you when you’re missing something they like.
As to adding points, what about research that shows the impact of adding and removing products? I’m not talking about a SKU here or a SKU there, but when you start talking about getting rid of brands, a vendor should be showing you good research as to why that’s a good idea.
Having clear guidelines helps the process. Answer the question, am I carrying this SKU for profit, revenue, to drive traffic or to satisfy impulse items? Second, is the category a statement category for me where breadth of assortment trumps sales?
It is a decision funnel and depends first on where the category falls and then the specific SKU. If a SKU is on the fence, use your gut as a merchant.
The biggest mistake is using sales per square foot as the sole criterion for cutting items. If a consumer is very loyal to an item, even when not purchased often, and that item is eliminated, there is a problem. You have to be able to identify your most valuable consumers and then examine their purchase history to ensure that important items are not eliminated.
There is an old study from IRI of a company that did eliminate a bunch of slow moving items to find that consumers loyal to a particular item went to another store to buy that item and spent $70 at the same time.
Whoever is responsible for the SKU rationalization project should not be responsible for any one particular category. If you miss cross-category product affinities, you can make a real mess.
I see that as the single most common problem – missing the forest for the trees.
One definition of “rationalize” is to think about or describe something in a way that explains it and makes it seem proper. Of course another is to find ways to waste less time, effort, and money.
Unfortunately, too many organizations approach SKU rationalization as pure science with rule-driven decisions based solely on product movement. These same operations then struggle to explain their actions to shoppers (and supplier partners) as being proper.
I support SKU rationalization, but I encourage retailers/wholesalers to do it rationally … with consumer satisfaction a vital part of the assortment decision.
Rule #1 in the article is key. I have always advised specific movement metrics. The exception is to carry a vital few unique items that have no substitutes. Other than that, I believe you should take your “gut” out of the decision process a bit more and leverage some of the tools available today to manage assortment.
The critical issue is understanding the market structure and decision tree, or more simply, what items compete most directly together, and belong in a bucket. If you don’t know that, you will delete the most unique and most profitable items and keep too many of the most common varieties.
There are so many factors to consider and it took a surprisingly long time for our industry to finally have software that truly brings the science to the art! Using loyalty card data to build a category decision tree and also measure incremental sales vs cannibalization and halo with science of “demand transference”…just as Ben Ball’s comment describes. While this is the flashy part…you also need to embed this science in a process flow that looks at the value of the baskets for each SKU and the value of customers who buy it.
Also, not only as average store sales but also individual stores often grouped into demographic or customer-centric store clusters because you may want to keep it in some stores and keep your planogram count manageable. A system that brings POS, loyalty, syndicated market data, demographics data sources together, doesn’t require deep analytical skills, and integrates directly to existing forecasting and supply chain systems.
These are good, common sense rules, but one needs to be added. The consumer substitution rule should be considered. The rule says if the item the customer wants is not available, they will buy a substitution item and there are no lost sales. In the real world it does not work that way. Each category has its own brand loyalty factor. With high loyalty, there are lost sales. If there’s low loyalty, there may be a substitution sale this time. What makes this rule difficult to work with is not only does the category factor in, but the consumer also has a loyalty factor.
All of the tips above speak to a category-centric view of the world. If you take a customer-centric view, you will ask a much more important question — what are the SKUs that my most valuable customers purchase. Those are the SKUs you MUST carry, even if they are less profitable to customers. You see, the research says that customers in general do not complain when they cannot find a product that they want — they just go away.
When those are your most valuable customers, the decision to cancel that SKU can cost a lot more than just the cost of the space.