PRICING NEWS
Same price everywhere in retail … the Handelsblatt and the discount lie
by Prof. Dr. Hans-Christian Riekhof
Are the prices for most foods the same at all retailers nationwide down to the cent?
One thing I must definitely preface these Pricing News with: I have valued the Handelsblatt as a quality medium for many years, and it has also developed very nicely in my humble opinion after the departure of Garbor Steingart. I particularly appreciate the many detailed interviews, for example with US economists, on important challenges of our time.
When Florian Kolf and Katrin Terpitz then title their contribution in the weekend edition of 01.-03.11.2024 with “The Discount Lie”, they have achieved their journalistic goal and gained my attention as a consumer and of course also as a pricing researcher. This is journalistic craftsmanship. So far so good.
Does Nutella cost the same everywhere?
Source: © iStockphoto
Vivid examples: Starting into pricing with anecdotal evidence
The authors start with anecdotal evidence: Nutella becomes more expensive at Lidl and Aldi Süd at the same time in May 2024! Is a scandal looming here? Further examples follow (which I will spare myself), and I quote verbatim:
“The examples are no coincidence, there is a system behind it. No matter which supermarket or discount store the customer goes to, the price for a product is usually identical down to the cent – and has been for years.”
Now it gets exciting: with Smhaggle to empirical evidence?
Of course, Florian Kolf and Katrin Terpitz provide the necessary empirical foundation for their rather daring thesis. They refer to a study commissioned by the Handelsblatt from the price comparison app Smhaggle and also let Smhaggle CEO Sven Reuter have his say in the text.
The empirical data base: what we know about retail prices …
In the almost double-page article, we learn
- that hundreds of products were checked in the Smhaggle study. However, ChatGPT says that an Edeka or Rewe store carries between 10,000 and 40,000 items. So can we say that the study maybe covers 1% of the assortment?
- that Smhaggle has more than 6 million receipts available as the basis for their analyses. Here too, ChatGPT tells us that Germans make about 20-22 million purchases in retail every day – every day, mind you. The 6 million receipts are cumulative data – over what period of time, we do not find out. But ChatGPT knows that the app officially launched in October 2019. If we estimate that Smhaggle has 10,000 receipts available per day (in relation to the daily 20 million purchases of Germans), then we see that we should view the empirical basis with considerable caution.
To use anecdotal evidence again: at the next opportunity I will visit Feinkost Meyer in Wennigstedt on Sylt and check what a 450-gram jar of Nutella costs there. And maybe also take a look at the regional champagne prices.
Champagne: Regionalization of products without regionalization of prices?
Source: © UCS
The empirical data basis: what we do not learn about retail prices …
With every empirical study, it is important to know WHO was surveyed, whether the study participants were selected at random, for example, or whether they were selected by another method so that they represent a meaningful selection, if not a representative one. In the case of Smhaggle, it can be assumed that there is a preselection: those who are particularly price-sensitive are more likely to install this app and use it. Do you, dear reader, have Smhaggle installed? So there is a self-selection error: I suspect that it is precisely the price-sensitive consumers who use this app. Read now: Pricing-Newsletter No. 40 (2020): The LSDC model – Pricing for Convenience Shoppers Whether the app users systematically differ from the average citizen or not, we unfortunately do not find out – but this could be easily represented. And what we also do not learn is the regional distribution of the data. Is there a north-south gradient – or an east-west gradient? My long-standing experience in retail pricing shows very clear regional price differentiations – based on the original data of the retailers, not on the basis of the data of some customers. Read now: Pricing-Newsletter No. 54 (2021): How does regional price differentiation work in practice? And also in drugstores, the prices are anything but uniform nationwide (apart from key products, of course). ChatGPT suspects that local competitors, rural vs. urban regions and also the costs for the store rents cause price differences, if not even enforce them.
So is there a discount lie?
We see that the empirical evidence is not very convincing. Therefore, Florian Kolf and Katrin Terpitz should be careful when they talk about a discount lie. The presented data certainly do not suggest this. My interpretation of the data would go in a completely different direction: there are a few hundred articles that customers actually have in mind and that they compare. And as a retailer, you should not stray too far from the competition with these customers. It is probably enough to empirically determine these lighthouse articles from the customer’s point of view and set a price here that does not make your own business look bad.
Individualized Pricing: How Digital Customer Cards Influence Prices
We learn little in the article by Florian Kolf and Katrin Terpitz about whether individual customers pay the same price for the jar of Nutella at all. Or whether they redeem a coupon at the checkout at Edeka or Rewe via the app. Maybe they also collect points via the retailer app, which are equivalent to a personal discount. An aspect that may become even more important in the future, as our case study from the DIY sector shows: Read now: Pricing-Newsletter No. 89 (2024): Individualized Pricing instead of flat discount: How location-based targeting at the POS works with app-based customer cards
Why media need to get better at dealing with numbers
What I wish for as a reader is a critical approach to numbers and empirical data. The methodological background of a study must at least be briefly illuminated, and there must be assessments of the validity of the study. Furthermore, I consider it essential not only to show snapshots, but especially to present long-term developments. And if the space allocated by the editorial team for the article is not sufficient: the many expert opinions of the article can be reduced in favor of usable empirical evidence.
What pricing experts can learn from this example
Regardless of the methodological questions I have just raised, it remains to be noted that the core idea of many pricing strategies, not only in retail, is to differentiate prices more strongly, both according to regional and temporal criteria
- and to underpin the price structure of the assortment with a strategically oriented and well-thought-out price architecture
- thereby increasing the complexity of pricing to the outside market – while it is reduced internally through appropriate systems and processes.
And that it is only conditionally sensible from a strategic point of view to align prices predominantly with the competition, I described in the Pricing-Newsletter No. 58 (2022): Read now: Pricing-Newsletter No. 58 (2022): When does it make sense to align pricing with the competition? Please let me know what experiences you have had in this area – and what tools you use to implement evidence-based pricing. Will we see each other in the new year at one of my Pricing Seminars in Hamburg? Best regards Your Prof. Hans-Christian Riekhof