Over the past ten years, online grocery has developed into my home turf. Ten years ago, online grocery was a relatively obscur industry. People knew, or assumed, it existed, but the vast majority of people had never tried it, let alone was using it on a regular basis. It was the time when companies like Oda and Picnic were in their infancy; you could see their potential, but these companies were so young that there wasn’t much too show for it yet.
It is a complicated industry from a logistical perspective, but also from a business perspective since the established brick-and-mortar chains are very good at defending their oligopoly (hello, dear antitrust authorities!), customer behavior is largely shaped by decades of physically going to stores, and significant capital is required to scale.
The most striking aspect of the industry, in my view, is something else entirely: the enormous gap between what is technically and operationally possible and what is actually being done in many places.
Living far away from any large city, my first experience as an online grocery customer was with a company called All You Need Fresh (now closed). Their service was delivery-only. Orders were shipped either directly from their warehouse (located in Bor in the Czech Republic, near the German border, so they could benefit from Eastern European labor cost while still serving German customers) or via DHL. At the time, AYNF had recently been acquired by DHL.
The people at AYNF were well aware that postal delivery was only a second-best solution. In fact, when I worked with them they said they’d like to give us some vouchers for us to try out their service, but not for the DHL delivery option. Since that was the only one available to me, living too far away from their warehouse, I tried it anyway. And that helped me understand right away why delivery by postal service really was not a great idea.
I believe the photos speak for themselves. (Fun Fact: You could return the empty cartons and all the trash for free at any DHL store; the closest DHL store for me was right inside an Edeka grocery store, which made it exceedingly clear that using this service wouldn’t make a lot of sense for me).
My team and I conducted a planning study for their warehouse. Interestingly, labor cost was not the decisive factor in their automation considerations. Labour availability in the border region was the real constraint; workers were scarce and highly sought after. Shortly after we submitted our proposal, DHL halted further investments and eventually sold the company. There are some reasons why I believe the business failed, but that is not the subject of this article.
In the years following my work with AYNF, I worked with very intensively with other amazing companies like Oda and Picnic, along with some major retailers and smaller national and international players. Oda and Picnic always stood out with everything they did, and to this day these two companies seem to be the benchmark for online grocery in Europe. But since I still live in the country side, I can’t use either of them. The only service available to me is curbside pickup from German retailer REWE. REWE, by the way, is the only traditional retailer that has pursued online grocery in a serious and sustained way.
Unfortunately, the REWE curb side experience is not great. Not because I have to physically drive to the store 10 km one-way — I happily accept that since it saves me a ton of time inside the store — but because there is still too much friction, some of which I have a hard time accepting.
One day, I wanted to pick up an order from the local REWE store. I had reserved a pickup window from 10:00 to 12:00. At 09:15, the app showed my order status as “Ready to pick up” (“Bestellung abholbereit” in the German app). I arrived at the store around 09:40 and asked for my order. The employee refused to hand it over, pointing out that my pickup window had not yet begun. My response was that the app clearly states that the order is READY TO PICK UP, which translates into IT CAN BE PICKED UP NOW. What followed was a back and forth for ten minutes before I could finally get my order. (A friend later told me he had exactly the same situation with the same guy at the same store, and he actually waited for 10 minutes until his pick up window officially started — which is something I most definitly will never accept).
Let’s disregard for a moment the fact that the store clerk was not exactly all-in on customer satisfaction. The more important part for me is: how can one of the largest grocery chains in Europe allow ambiguity of this kind to persist in their user interface? “Ready to pick up” to me — and I care about precise language — sounds a lot like “come and get your order, if you want, even if you are a little ahead of your reserved time window”. “Ready to pick up” implies availability, not conditional availability.
Another point of friction is when there is a problem with some of the products you ordered. After one order, I discovered that several vegetables were in very poor condition. I looked for an option in the app to report damaged items and request a refund. There was none. The app directed me to customer service; the website suggested contacting the store.
So I called. On the phone, I was told to return the items in person. I told them I am not going to drive 20 km to return three rotten tomatoes, and suggest that I’d document the damage with some pictures instead and claim the refund next time I’m at the store. They said that’s ok.
A week later at the store, when I picked up my next order, I said I’d like to claim a refund for some damaged items from my previous order, and that I had taken some photos to document the damage. The store clerk told me that I should have returned the items to the store right away, and THAT HE CANNOT REFUND ITEMS BASED ON PHOTOS. Well, that was the same guy who complained about me arriving ahead of my pick up time window earlier. I asked him to get me his manager, and his manager accepted the refund without much discussion.
Again, the point is not individual behaviour. The real question is systemic: how can an online ordering process exist without a standardized, digital workflow IN THE APP for reporting damaged items and issuing refunds? How can PHYSICALLY BRINGING THE ITEMS BACK TO THE STORE be the proposed solution? The reason I order online is precisely the same reason why driving back to the store to claim a refund for items worth three Euros is not an option: because I don’t have much time to waste. Requiring customers to drive back to the store defeats that purpose entirely. How is that not obvious to EVERYONE at REWE in charge of processes and the app for online orders? They are processing thousands of orders per day. Is this a new issue that nobody ever raised before? Are my expectations somehow out of place? REWE has (or had, as some have left) some excellent people in their fulfillment and automation team, and they are effectively the only German chain who took the online business serious early on; did they run out of money when hiring process and IT people?
REWE’s biggest competitor in Germany is Edeka, a food retail chain of similar size with the same portfolio, serving the same customer segments. In contrast to REWE, Edeka really did miss the train for online grocery entirely. Our two local Edeka store, pretty sizable by German standards, advertise home delivery. Unfortunately, I could not find out how to order since they don’t have a website.
I finally learnt you need to call them. Of course, I am not going to do that.
You know, when I was a little child in the early 1990s, there was a SPAR store in my hometown. I remember that my mother ordered for home delivery via fax sometimes. This worked. Even fax is a much better idea than CALLING and listing a hundred different items. And that was understood more than 30 years ago.
A couple of weeks ago on the way back from a visit to the Christmas market with my daughter in a neighboring town, I walked by another Edeka store. There, I saw this sign:
Amid the existence of companies like Oda and Picnic, Edeka has come to the point where they accept orders via telephone or email on Tuesday, 09:00 – 11:00, as well as Wednesday, 09:00 – 11:00. Also, deliveries cannot be scheduled “for logistical reasons”. And preferably you should not order more than beverage crate (the reason they don’t like it is the very same reason why customers would want to order beverage crates: they are heavy).
I am sure that nobody else in that place who looked at this sign took any issues with it. Having worked with online grocery for about ten years, and with some of the most amazing companies in the industry, it is fascinating to see the gap between what CAN BE DONE and what IS BEING DONE elsewhere. The contrast is a bit like the comparison of the software in my Volkswagen Passat (with VW-developed software) and a Tesla. It is just.. you know… there is no comparison.
Which brings me to a broader point: there is SUCH a big and persistent gap in logistics between what CAN BE DONE and what IS BEING DONE in MOST places. A small number of companies are cutting edge in their logistics operations, but most are.. not. Many are stuck with problems for which effective solutions have existed for decades. And I want to make it very clear that this is NOT about the adoption of modern technologies. In fact, I am familiar with quite a few companies that have adopted automation technology and/or fancy software, and they are still failing. Moreover, “everybody” has an AutoStore system these days, but every other system I visit runs like crap (more about that in a separate article). There are lots of low-hanging fruits that can be identified through simple audits / warehouse health check-ups. I have been running plenty of simple health check-ups in warehouses across Europe, and I don’t think there has been a single warehouse visit where we did not identify some quick and very practical opportunities to improve overall performance. Some audits are followed up with some straight and simple data analyses that reveal, and quantify even more potential (especially in the common “our warehouse is full” cases).
Part of the problem that I illustrated with above examples certainly is the well-known Knowing-Doing Gap (Pfeffer & Sutton): the information to act is available, but nobody follows through. This behavior, in turn, is at least partly a function of culture, one of the least acknowledged yet most important aspects of management. The bigger part, however, seems to be a gap in awareness, which is good news, because that’s much easier to fix.
