MMI Brands is in a strategic alliance with IRI, a global leader in syndicated sales data and market research, to develop new and emerging brands introduction initiatives.
Robert Berg has been working in the CPG industry for more than 25 years. After earning an MBA from The Wharton School, Robert began his career at Coopers & Lybrand Consulting and PriceWaterhouseCoopers Consulting. Working for IBM Global Business Services, Robert worked on one of the largest CPG mergers in 2001. During his 13-year tenure with IRI, Robert has focused on rolling out new solutions for both CPG manufacturers and retailers and is currently working with mid-market clients. He is also charged with helping IRI clients receive the greatest value from their investment in IRI’s consumer and shopper solutions.
- We have gotten a lot of responses from your last interview from people who want to get into more specifics concerning store-level insights. They want to see some specifics. Would you be able to share an example?
Absolutely, I would be happy to. Previously, I discussed two questions concerning retailer insights, which we can take to the store-level. Those two questions were:
- At which retailers or stores do they shop at higher concentration (even if your products are not offered there currently) or how important is your shopper to the retailer’s sales?
- At which retailers do you have a reasonable opportunity to make the most money this year?
Allow me to explain this with a picture of the measures and a graphic on the results. For the first question, we calculate how much each household (HH) spends at a retailer and then sum the share of wallet of those HH’s we identified as the target.
In this example, we created a target for a new emerging product and generated this information at the banner level. (I’ll show store level in the next example.) We are going to see where are my shopper’s spending their money in higher concentration and how important are they to the retailer?
How do you read this chart? Let me give you the definition of all the measures and then we can discuss.
Total Estimated All Commodity Volume (ACV) (MM) is the sales (of the items IRI tracks) of that store. This measure gives one a sense of the size of the retailer or store.
Target ACV (MM) is how much of the retailer’s total estimated all-commodity volume (ACV) is accounted by our target shoppers.
Target % ACV is (Target ACV/ Total Estimated ACV) is the percentage of the total retailer ACV$ our target shopper accounts for.
Target % ACV Index is (Target % ACV/ Average Target % ACV) indicates whether our target % ACV is above or below the average Target % ACV reference. The reference we use in this analysis could be the average U.S. food banner.
In this example, there are a few Albertsons’ local banner stores that are of great interest to us. In Lucky, Pavilion’s and Randall’s, our target shopper accounts for almost 50% or more of the estimated ACV of these banners. That equates to every $10 that is spent at these banners, our target shopper is contributing about $5. For a new emerging product, it’s really important that we find its loyal consumers as efficiently and effectively as possible. This method finds high propensity HHs without having to distribute or sell the product. We are applying a consistent approach across all retailers using HH actual purchasing (buying a product) and shopping (spending at a retailer) to inform the results.
This information allows us to identify where we can place the product — in retailers that help them build loyalty and the right assortment to important shoppers. As a standalone measure, manufacturers and retailers use target % ACV to help identify where to have secondary placements, in-store events and initial roll-outs.
The next set of measures we are going to look at addresses at which retailers is there a reasonable opportunity to make the most money this year. Again, let’s look at the measures and then provide some explanation. In this example, I’ll go to the store-level.
Again, allow me to provide the definition of all the measures and then we can discuss.
Household Index: A measure of the store size based on the share of wallet a HH accounts for from that retailer and the number of households in the trade area. One group of HHs can spend $1,000 per year of their share of wallet at Costco and another group of HHs spends $4,000 per year. It will take four-times as many of the first group of HHs to equal one HH in the second group. Sum up all the HHs shopping at the Costco’s and divide by a reference — in this case, the average club Store. In this example, the Costco located at 4502 East Oak St. in Phoenix, Arizona, is 34% larger in size of HHs than the average club store. This value is not product specific. As long as we use the same reference, it will be the same value.
Profile Alignment Index is the potency of the product’s sales per HH for the store to the average sales per HH in the reference banner. This measure indicates whether the average shopper in the store has a propensity to buy more or less of the product than the average shopper (in this instance, the average club shopper). This measure is product specific. In this example, the average Costco shopper at 4502 East Oak St. in Phoenix, Arizona, has the propensity to buy 23% more product than the average club shopper. This is very important for decisions. Even if the HH Index is low, higher than average profile alignment index stores are great opportunities for new emerging products.
Product Potential Index is the comparison of what the product should sell in this banner relative to the average club store, given the same conditions for this product over the last 52-week time period. One can see that size of the HH index times the potency of the profile alignment index will equal the product potential. The Costco at 4502 East Oak St. in Phoenix, Arizona has size (134 Index) and potency (123 Index), therefore its product potential is 166. We really need to get this product in that store. However, there are two other Costco stores where the product potential is even higher, due to the size of the stores.
Product Potential Dollars is the dollar amount the product should sell in this banner given sales and sales conditions over the last 52 weeks at club. We use proxies for the new emerging product to provide a sense of the magnitude for a store’s sales expectation without using it as a predictive measure.
This bears repeating from the last interview. Is this measure predictive? Product potential is forward-looking, but it is not predictive. We use past behavior to provide a consistent value across the retail locations and products IRI tracks. The information accounts for the last 52 weeks of pricing, price differential and support. What is important to understand about this product potential index and product potential dollar value is for sales, marketing, and operational purposes, not for investment purposes. IRI can provide predictive modeling and new product simulation tools but that’s a different solution and requires different inputs. Let’s leave that for a different conversation.
Manufacturers and retailers use this measure to understand where to place the product for best sell-through. Let’s put these measures together in what we call the activation grid and understand how it helps lead to different store prioritization strategies and tactics.
In the store activation grid, the target % ACV index is on the y-axis and the product potential index is on the x-axis. We recommend that you focus on banners and stores that reside in the upper right quadrant. It is important to do for existing products and even more so for new emerging products. Minimize what you spend on stores in the lower-left quadrant. Be smart about how you spend on distributing your product in the upper-left and lower-right quadrants. Build your brand and retailer relationships for the right reasons in the upper-right quadrant.
We all talk about risk management and focusing on the areas that make the most difference. Time and time again, I listen to companies discuss how they have to be in a particular retailer. Too often that retailer is good, but isn’t driving sufficient sales or that location turns out to be inappropriate for the product. I’m getting pulled in at that point because now the risk of de-listment or the specter of poor performance necessitates an extraordinary effort.
The process that I go through is the very similar to what I have laid out here. There’s no reason to “roll the dice” on whether your retailer prioritization and their store roll-out is going to more or less successful. We can effectively and efficiently manage that risk from the beginning.
Let me end with an example. A client had a new product for a premium baked item. They approached a retailer who like the product but didn’t believe it “fit” their shopper and turned it down. The retailer was correct. It didn’t fit their “average” shopper; however, we found that 20% of their locations had a large concentration of the target shopper and higher levels of product potential. The retailer recognized that these locations didn’t follow the average store sales. That product generated $6 million in 12 months at that retailer. This was one product at one retailer; it did not need to be in all of the stores.
Be strong where your biggest opportunities are and don’t expend your precious resources — money, time or effort — on the places where opportunity are not apparent at the moment. There are enough places and revenue in places where the data is indicating to go than to take an undue risk.