In established categories or categories with commodity items, sales rates (or gross margin contribution) make perfect sense. In new or emerging categories I think it’s a little trickier. It may take a certain number of SKUs just to tell the story. Some SKUs may hit the threshold, some may not, but the story must be told. If the offering is brand new, sufficient time must be allowed to ensure the customer has actually become aware that the new product is available. Over time, pure $$$ can make the call. But some level of incubation and nurturing is going to be warranted for new and emerging stories and items.

Using sales-based hurdle rates has never been the best to determine item retention, although I recognize that many companies aren’t sophisticated enough to understand all the ways that a product can contribute to the store. An item that fits a local niche, increases the perception of variety in a narrow assortment, that helps set price points in a category (i.e. high or low priced products) can all be useful beyond their sales numbers. And the argument that what space they take up could be better used is weak – the 80/20 rule will have covered that.

For many of the categories mentioned in the article; private label, organic, etc., this makes sense. The challenge will be on where to draw the line. It’s a slippery slope when there are so many exceptions to the rule. I agree that there should be exceptions and that additional metrics should be considered, but I can see many suppliers trying to make the case for why their products deserve more time or that they are now becoming trendy or whatever reason or excuse they may come up with for staying on the shelf longer.

Merchandising/shelf management has always been a combination of science (data) and art so a few more nuances can’t hurt, can they?

Hurdle rates alone are no longer anywhere close to good enough measurements for assortment planning. In my opinion, proper lineup development requires a sophisticated balance of sales, growth, competitive positioning, basket analysis and human instinct. In today’s fragmented and niche-driven economy, one size clearly does not fit all …

I don’t believe that using hurdle rates alone makes sense. In the book industry, we would stock one copy of some titles because they consistently sold an average of one copy a month. That title kept some customers coming to the stores. The answer to this quandary lies in what the retailer believes is the overriding reason behind the decision to stock or not to stock some slow items: to offer some modicum of variety, or to maximize sales per linear foot. Pick it!

SKU rationalization used to be a simple numbers game. If an item didn’t meet the hurdle rate for sales (units and/or dollars) or gross margin it was cut. Today it still is a numbers game but there are far more data points to be considered. As the article points out, one is the role of the category and of the items being evaluated. An item may be new or you may have to carry it because your contract with the supplier requires you carry X number of their SKUs and it may be the best of the worst. Another could be the item’s impact on market baskets. The item my not be a great seller or carry a larger gross margin but it may be that those who buy that specific item also have large or even very large market baskets. They may be coming to your location because you are the only retailer that carries that slow-moving item. Cut too deep and sales decline as many retailers including Walmart have found out. Once the… Read more »

First of all, let’s be realistic. Many SKUs get to — and stay — on the shelves because of trade funds, not actual consumer demand. Secondly — outside of retailers such as Aldi, Trader Joe’s and a handful of others — most stores are still carrying way too much inventory because they either want those trade dollars or they really don’t know their customers well enough to know what to kill and what to keep. Velocity reports/hurdle analysis is one way to know what to keep stocking but it doesn’t help you figure out what you should be stocking that you currently aren’t. Sol Price, the founder of Price Club, used to talk about “the intelligent loss of business,” i.e., not stocking “dog” items and letting the customer cherry pick the competition. So hurdles are one tool — and a valuable one — but not the only tool retailers should rely on. Great retails like H-E-B, Wegmans and may luxury branders know two things some retailers don’t — who they are and how they want… Read more »

Sales-based hurdle rates only provide one cut at the decision process for carrying goods in a store. As Walmart has decided, there may be other variables at play that are equally or more important for different reason. When a new product category to stock is decided upon, it will have a low hurdle rate for some time. Using sales-based hurdle rates only puts this category at a shelving disadvantage to the point the category may never get its fair shake eliminating that category from the retailer. This may be true as the retailer attempts to move the store’s focus to other areas such as gluten-free or private labels. Some categories may never make a pure sales-based hurdle rate but are important to attract consumers to the stores. Having more decision variables makes the category decision process more complex but will result in better store strategies over time. The real measure is total store sales growing.

While in business school, I interviewed for a summer internship with one of the elite consulting boutiques. In the case study portion, the key factor revolved around dollar sales per square foot as the magic lever to measure future success of a retailer. Granted this was several decades ago, but time has not diminished the ubiquity or use of this sales-based metric. Such performance and productivity measures can be highly misleading if taken to extremes as when they become the sole driver for creating assortments. Localization matters, trends matter, newness matters, variety matters just as does having the most popular and in-demand products. If all competitors focus on sales-based hurdle rates, their respective stores will be undifferentiated on assortment and will rely solely on pricing and visual merchandising. Such a scenario also drives consumers to more online sales to satisfy their unmet demands. Sales-based metrics belong in the toolkit, they just aren’t the only ones and certainly don’t complete the set. Today’s experience economy requires assortments to be part of a larger engagement story that… Read more »

This is a difficult issue for a supermarket. When you strip it down to the issue of what the supermarket business is, you get down to the fixed cost of operating the business, the limited real estate that can be utilized and the ROI on that limited real estate. That would suggest that the only measures are movement and margins.

But then you look in a typical shopper’s basket. How do you satisfy the shopper that has, say, 10 percent of their picks that might not meet those metrics? Eliminate those items and you may eliminate the shopper all together.

Great merchandising requires human instinct. No strict reliance on metrics (whether adjusted per the gluten-free products at Walmart or not) will replace that — and build a strong long-term relationship with consumers.

We need always remember Goodhart’s Law: Once a metric becomes a goal (of ANY type), it’s not longer a metric that’s useful for managing stores.

Why the human instinct? Reliance on metrics has CREATED a major part of the store experience problem. Only people can see subtle emotional connections that products make — like how exceptionally interesting (although possibly unprofitable) products DRAW consumers to the store and make their shopping satisfying.

The key remains: Trust merchandisers to put the right product on the shelf and manage the total departmental sales.

Hypothetically the extreme would be the end of loss-leaders, bundled products, and cross-sell/upsell opportunities. Clearly stores have not entirely embraced assortment selection based on historical sales and, as many have mentioned, there may be competing goals that are strategic such as entering new markets, capturing market share, delivering longer-term sales results, or encouraging non-variable spend.

The interesting thing about hurdle rates is that when it comes to online buying, similar techniques are exactly where retailers begin — historical sales data: what’s selling the most. Retailers then expand and optimize their selling and merch. Models of data are plentiful — especially cross correlation about the combination of products that are selling the most. As retailers begin collecting more data in the store beyond POS logs, they’ll begin building models that can capture interesting and profitable opportunities through test and learn. Competitors who cannot/will not do this will be left behind.

Oddly enough, in past trials when store managers were permitted to exercise some degrees of freedom in selecting product and setting up an end cap, results tended to be superior. There are some things that the raw numbers may not reveal. Without context, perhaps understanding which customer “would buy” a product as opposed to just how much of what you selected they did buy, still requires some merchant and local touch.

Hurdles are important but not the only criterion for making a stay or go decision. What purpose does the item serve for which group of consumers? Are those consumers important to you? Do you want to support the purpose that item serves for those consumers? The go/stay decision needs to consider a variety of factors, only one of which is a numeric hurdle.

From the early days of the Category Management movement, when “rank-and-cut” assortment management became a thing, there was an awareness that those methods were too crude. It’s not just about what items sell most — it’s about what items are important to high-value shoppers? Fold in the trade marketing incentives (as others have mentioned here) and you have a multi-variate challenge.
The calculus has often been made harder by poor on-shelf availability, which can skew the metrics, and generally lax in-store implementation.

All three of the experts cited in this article are long-time, highly-astute observers of this phenomenon, and their suggestions should be heeded.

One key to proper product assortment is knowing who your customers are. Are more Hispanics shopping the store because more are living in the neighborhood? If so, the assortments need to change to cater to these new local customers. Shopper demographics always change, and grocers need to monitor these changes so they can give customers what they want to buy.