Dairy Marketing Practice | Contributing Columnist


On Empty Shelves and Other Follies

Dan Strongin ASQ CMQ/OE Uncorporate Consultant

February 9, 2018


Last month the Business Insider published a story about Whole Foods shelves being empty. Everyone was quick to blame Amazon, but as it turns out, it was caused by the unintended and unexpected effects of adopting a new approach to product and shelf management, according to an article posted on the website of Supermarket News, on January 23rd

In that post, the writer did her best to explain what this new system was. Unless you are a process wonk, it can be hard to get it right. And, once again, her sources “blame” Toyota, and their process for the delivery of parts to the assembly line called “Just in Time.”
People in other industries have been obsessively copying Toyota for years, without really understanding what they do, so why not Whole Foods ( and, it seems, Target)? Whole Foods called it their “order to shelf” initiative.

It’s easy to see why not having to pay for or maintain any inventory would be seductive. After all, the suppliers for Toyota seem to manage to deliver products just in time, within a few minutes of when they will be installed, cleverly eliminating the necessity for Toyota to carry inventory, or pay for products until they’re used.

Another way to put this is Toyota has unloaded its risks onto the shoulders of their suppliers. Nice trick, if you can get away with it. Many suppliers for Toyota, who deliver Just in Time, are small companies contracted to work for Toyota exclusively. Toyota has the muscle to require these suppliers follow their scheme.

Moreover, a parts supplier or an automobile factory are completely different from supermarkets and retail stores. Without going into the gory details, a supermarket deals directly with those fickle things we call consumers, and has more variability and more products; thus, more chance for wiggly things to happen. Using an on-demand, or just in time system, requires some measure of predictability, and needs to take into account more than just what sold yesterday.

I can see it in my mind’s eye. A bunch of MBAs at the corporate offices all excited about the chance to apply the statistics that they learned badly in college, (not realizing that the kind of statistics that they learned in college is useless for prediction.)

Which is not to say that the goal of lowering inventory is not a valuable goal. But, using your buying muscle to force your suppliers to take on your risks without proper compensation, is a bit “gangsta” in my opinion. In addition to lacking the leverage Toyota has over suppliers who only work for them, supermarkets like Whole Foods have suppliers who have other customers.

So there is a kind of hubris involved in the application of this program to a completely different industry than what it was designed for. Did they consult with experienced experts in Supply Chain Management? There are people who have the proper training, who do know analytic statistics, that could have devised a system that would give them a more manageable inventory without the dangerous unintended consequences. Customers are fickle and have little patience. They will abandon you for something as little as the price of a piece of Brie was marked wrong; imagine empty shelves.

This is a problem of the age of Google, lots of information, little knowledge and experience behind it. A good solution could be developed either using the age-old system of buffering supply against the variations in daily and weekly demand, what used to be called and on hand build to system, or a par system. (I was brought up on it.)

Or, by using a much more sophisticated and incredibly more accurate system that would account for the variations in product movement over time, predicting within certain limits a range of potential ups and downs, ensuring many fewer products missing on the shelves and much less inventory to manage. In other words, more money in the pocket for management.

It seems like someone in upper management read an article or a discussion post on LinkedIn, and we’re dumb enough to apply without really having an understanding of what they were doing, or the common sense to test the thing in small ways first before rolling it out into the whole shebang. Another example of, as Bob Lutz wrote in Car Guys vs. Bean Counters, “Business Schools have produced generations of number-crunching, alternate-scenario-loving, spreadsheet-addicted idiot-savants.”

The irony is that the creator of the Toyota production system, Taiichi Ohno, got his initial ideas from a visit to an American supermarket in the 1950s. However, Toyota does not and never has worked on a pure “on demand” or just in time model for deciding on how much product they should produce.

They work on a modified version, using sales projections from the different divisions built from data collected over time to predict how many cars to make in a month: statistical models based on pragmatic, analytic, not ordinary count em up statistics, regression analysis, and probabilistic equations be damned!

Sitting around the executive offices, looking at spreadsheet profit, cooking up hair-brained schemes from what they learned in Buness School, rather than having the humility to go to the stores
and the talk to the many hundreds of dedicated willing workers who deal with these problems on a daily basis within their own company is most likely the problem. They kind of clevered themselves into a kerfuffle.

I do know some people who worked with Amazon on their predictive statistics. Ironically, if anyone is prepared to improve the situation, it is Amazon. Though I do still believe they would be better served by involving the people who work for them and are closer to the customers and the problem. A store is not a website. After all, it was the employees who brought all the great cheeses into the stores in the first place. DS


The views expressed above do not necessarily reflect those of Cheese Reporter.



Dan Strongin

Dan Strongin is a former president of the American Cheese Society, chef and business coach for small to medium value added businesses, and the owner of the sites learn.managenaturally.com, and the Facebook group Enjoy Cheese. His online course: “Cheese: How to Buy, Store, Taste, Pair, Talk About and Serve”, is available at enjoycheese.net. Dan can be reached via email at dan@danstrongin.com.

Recent Columns

Our Friend - Dan Carter
December 29, 2017

Artisan Cheese, The FDA, And The New Administration
October 21, 2017

Time's Winged Chariot Goes Boom
July 28, 2017

Brazilian Cheese Makers Break The Law, Win Awards
June 30, 2017

Safe And Sane Food Policy:
Fix The Causes To Fix The Results
December 16, 2016

Sane, Safe & Effective Food Policy
October 28, 2016

A Ray Of Hope For A More Effective Food Safety Policy
September 30, 2016

In Her Own Words - Linnea Burnham
August 19, 2016

ACS- 32 & Aging Well
July 22, 2016

What A Wonderful Life: A Tribute to Steve Ehlers
June 17, 2016

What do you think about 
Dan Strongin's Comments?*

Please tell us if you are a
Dairy product manufacturer 
Dairy marketer /importer/exporter
Milk producer
Supplier to manufacturer