Which positions at Allegro do you think, when you left, had to improve? Which roles needed some sort of improvement from the day you left?

What I did in Amazon, and where I think Allegro's biggest challenge is, I said: headquarters is Luxembourg for Europe. If you want a big job, then you need to come to Luxembourg. If you don't want to come to Luxembourg, that's fine, I'm not going to fire you. But most of the responsibility eventually is going to move to Luxembourg. That's what we did. That kept cost down and actually gave us considerable more efficiency because instead of running five national companies, we ran a single entity across Europe. That's for Amazon, and I think that's true for Allegro. If they want to scale in countries, remember that all of these countries are much smaller than the original core of Amazon. Poland is the biggest country in Central Europe, and all the other countries that they're going into, excluding Slovenia with its 1.5 million, Czech has 10 million, Slovakia has 6 million. Slovakia, as a country, is smaller than the city I was born in. You cannot create a national organization for the city of Los Angeles. You wouldn't. You continue to run it as part of the American organization. That's also what we did. You won't develop a website specifically for Los Angeles or Slovakia or the Czech Republic. It's one website. Everything in the website is the same for everywhere, except for language, payments, and ship options. That's what, by the way, companies like Temu do.

Which positions at Allegro do you think, when you left, had to improve? Which roles needed some sort of improvement from the day you left?

I think that is still, for what is fundamentally a very lethargic national organization, a difficult change in mindset. One of my biggest frustrations when I was running there is: "Well, the Slovaks are like that." And it's like, all right, well, how long have you worked in Slovakia? Do you speak Slovakian? How long have you spent in the Slovakian market? The answer was always not very little, none at all. That is a problem. If they don't get a handle on that, and I'm confident, at least from tech perspective, it's hardwired now, it's going to be difficult to win in those small markets because that's the reason why Amazon doesn't go in. Amazon doesn't go in because it costs them $100 million every time they stand up a new website.

If you own the supermarket, you have everything centralized, so that would make more economic sense rather than a 3P model, probably.

Yes. You have to figure out how to make it work. There are some very good retail operators. Dino is a bit old-fashioned and does not have full coverage, but I think they're pretty well run. If you fulfill from a supermarket and you do not have really solid planograms and really good on-shelf availability, your business is going to fall apart quickly. The capex to build dark stores is quite high when your volume is low. You need to find some way to build a viable grocery business that works at low volumes. I cannot see that working on anything other than a store-based picking model, which requires extremely well-run supermarkets, which not everyone has.