Just Eat is leaving Denmark on 30 April 2026.
The most obvious story is to ask who gets the customers.
Uber Eats is already mentioned in Just Eat Takeaway.com's own announcement through a marketing agreement offering customers the chance to switch platforms. That's a normal market story: one player exits, another stands ready, and the traffic is nudged along.
But that's not the most important question.
The most important question is this:
Why can a restaurant's digital existence disappear with a platform at all?
If a restaurant loses its visibility, its customer access, its order path, and its digital relationship when a private gateway exits the market — then it wasn't just an app that closed.
The connection was never really the restaurant's to begin with.
Switching platforms doesn't solve the ownership problem
It's easy to understand why platforms exist.
They reduce friction.
They aggregate restaurants.
They make payment, delivery, ranking, customer service, and reviews easier to handle in a single interface.
That's a real service.
The problem starts when the service slides into ownership of the relationship.
A platform helps the customer find the restaurant.
Shortly after, it also owns the contact path.
Then it owns the order flow.
Then it owns the review history.
Then it owns the customer data.
Then the restaurant's exit becomes expensive.
No single step necessarily looks dramatic.
But the sum does.
When the platform leaves the market, it becomes visible how much of the connection was held outside the restaurant itself.
That's why Just Eat's exit from Denmark is interesting beyond the ordinary business story. When a major platform leaves a market, it's fair to ask whether the next chapter should simply be another private passage.
The door is not the house
The distinction I'm trying to hold is simple:
Helping two parties find each other should not automatically grant the right to own the relationship afterwards.
Discovery is a door.
It should not automatically become the house.
That's the idea Pizza4People is built around.
Pizza4People is not a new delivery app.
It is not a finished restaurant platform.
It is not a production-ready alternative to Just Eat, Wolt, or Uber Eats.
It is an open protocol proof for a different distribution of responsibility.
The customer can find the restaurant through a registry.
But the menu lives at the restaurant's node.
The order endpoint lives at the restaurant's node.
The identity key lives at the restaurant's node.
The restaurant's operator decides whether it is menu-only, test-order, or live-order.
The registry helps the customer find the door.
It does not become the restaurant.
What the protocol is actually trying to prove
The current claim is narrow.
That matters.
Pizza4People is not trying to prove that the entire restaurant economy can be replaced by a weekend protocol.
It is trying to prove something more fundamental:
that discovery, menu access, and order submission can be separated from platform ownership.
In that model there are three base roles.
The client asks for restaurants.
The registry responds with public node metadata.
The restaurant's node owns the menu, status, order flow, and operator control.
After discovery, the client should speak directly to the node.
That means the registry does not need to be:
- order broker
- menu host
- payment processor
- review gatekeeper
- customer account owner
If it becomes those things, you're just building a new platform under a different name.
That's the collapse Pizza4People is trying to avoid at the protocol boundary.
Commercial services are welcome
This is not an argument against companies.
It is not an argument against profit, delivery, payment, hosting, POS integrations, printers, trust services, or restaurant support.
On the contrary.
If the base layer is open, more companies can compete on top of it.
A payment provider can provide payment.
A delivery provider can provide delivery.
A local operator can help restaurants with operations.
A trust provider can handle verification.
A POS provider can handle integration.
But none of them should automatically own the connection between restaurant and customer.
That's the difference.
The market doesn't become less commercial because the base layer is open.
It can become more healthily competitive, because more players can build modules without first having to own the entire relationship.
Readable markets
This is also why Pizza4People isn't only about pizza.
Pizza is just a good first laboratory, because it's concrete.
Everyone understands a pizza order.
Everyone understands a restaurant.
Everyone understands that a customer needs to find a menu and submit an order.
But behind that simple action lie many layers:
- discovery
- ranking
- menu
- payment
- delivery
- trust
- reviews
- data
- exit
When all those layers are hidden inside a single private platform, the market looks simpler than it is.
The customer sees an interface.
The restaurant sees a sales channel.
But neither can necessarily read the infrastructure.
A readable market is one where the important layers can be seen, separated, verified, and replaced.
Not everything needs to be visible all the time.
Not every customer needs to read a technical manifesto before dinner.
But the layers should be examinable when it matters.
And exit should not mean the entire digital relationship collapses.
What's missing
It's equally important to say what Pizza4People is not.
It is not a finished restaurant network.
It has not production-tested payment.
It has no delivery.
It has no CVR-bound trust.
It has no support, refunds, dispute handling, or broad restaurant onboarding.
The first real restaurant pilot is still the next step.
The current proof is public, but that is not the same as production.
That boundary needs to stay clean.
Otherwise the argument gets weaker, not stronger.
Because the point is precisely not to hide the gaps behind launch language.
The point is to make a small piece of market infrastructure legible enough that it can be attacked, tested, copied, or rejected.
The open question
When Just Eat closes in Denmark, you can choose to read it as a competitor story.
One platform out.
Another platform in.
That's the quick read.
The slower read is more interesting:
If restaurants and customers lose their connection because a platform exits the market, what does that say about the market's architecture?
Maybe the next layer shouldn't just be a new owner of the same passage.
Maybe the base connection should be more open.
Not because all platforms are bad.
But because access should not become ownership without anyone noticing.
Pizza4People is a small proof of that claim.
Nothing more than that.
But nothing less either.
— Dennis Hedegreen
The Danish version came first.
That was a mistake.
The draft was in Danish. I published the draft.
The Danish version is still up: Når en platform forlader markedet.
— Dennis Hedegreen, corrected publicly
