Do you need storefront monitoring if you already have Deliverect or UrbanPiper?
Restaurant chains running an integration platform already have something watching, and the honest answer is that it watches a different object. An integration reports the health of the connection between your systems and the platform. The storefront is separate, and it can be switched off by the platform, by an automatic rule inside the platform, or by a staff member on a tablet, with none of those events travelling through the connection. Both measurements are useful. Neither replaces the other.
What does an integration actually see?
It sees itself. An integration platform holds the link between a point of sale and a marketplace, pushes menus one way and orders the other, and reports whether that link is working. When it says the store is fine, it is telling the truth about the thing in its field of view. The field of view is the limit. Ask what the customer sees and the integration cannot answer, because the customer is not looking at the connection. The customer is looking at a card in an app, and that card has its own state, set by the platform.
What can close a storefront without touching the integration?
Most of the mechanisms, once you list them by platform.
A platform can close a store on its own account. Careem has a state called “Outlet Closed” that only Careem can lift. Deliveroo has a “Forced Closure” that the partner cannot reverse. Keeta has a Platform Override that removes delivery while leaving the store open for collection. An automatic rule can close it. Deliveroo closes a site after three automatic rejections inside fifteen consecutive minutes. Just Eat takes the storefront down after a single unaccepted order. DoorDash deactivates after roughly five minutes of tablet silence. Talabat has a check in feature where a shop scheduled to open stays closed if nobody acknowledges the opening within thirty minutes.
A person can close it. Any staff member with the partner tablet can pause orders during a rush and forget to unpause them.
The connection between your point of sale and the platform is intact through every one of those. The integration has nothing to report because nothing happened to the integration.
Does an integration platform ever admit this?
Sometimes, and in writing. Deliverect states in its own help centre that “Deliverect can no longer open or close your store on Jahez. Any request to open or close your store must be made directly with Jahez”, and adds that visibility status must be set back by the platform itself. That is a clear statement of where the boundary of an integration lies on one platform.
Read it as a description rather than as a criticism. It is exactly what an integration is: a connection with a defined reach, honestly documented.
What does the outside in measurement add?
It reads the public listing, which is the object the customer actually interacts with. That captures every cause of a dark storefront rather than the subset that flows through one connection, and it works identically on platforms where the chain has no integration at all. It also captures things that are not availability. What rating is displayed on the card right now, whether the promotion a chain is paying for is visible to a customer at a given address, whether the price on the app matches the price that was set, where the branch sits in the platform’s own list, and whether the platform will deliver to a given street.
None of those are integration questions. All of them are storefront questions.
What does the integration do better?
It acts. Tools that sit inside the stack can push a store back open automatically once the blocking condition clears, and on platforms where that is permitted it closes the loop without a human. Outside in monitoring reports and stops there. It also owns the menu. If prices and items on a platform are wrong because the source of truth is wrong, the fix belongs in the system that publishes the menu, not in the system that reads the storefront.
And it moves the orders. That is the primary job and nothing about storefront monitoring touches it.
So which one does a chain need?
Both, for different reasons, and the overlap is smaller than it looks. A chain with an integration and no storefront monitoring is confident and partially blind: it knows its plumbing is sound and does not know what a customer in a given district is being shown. A chain with storefront monitoring and no integration knows exactly what customers see and has to act on it by hand.
The useful test is a question. When a branch stops taking orders on one platform for three hours in the middle of dinner service, which system tells you, and how long does it take? If the answer is that a franchisee phoned the next morning, the gap is not in the integration.
How much does this matter in practice?
Enough to be measurable. Interruptions on the platforms we watch in the UAE run to about three hours each on average, and the ones that start after 22:00 tend to be the long ones because nobody is looking. Across a full month the average listing loses close to a full trading day. In our July 2026 UAE panel that was 9.5 hours per listing.
Multiply by the number of platforms a chain sells on and the figure stops being a rounding error, and none of it appears in an integration report.
Kitchain (kitchain.co) is the outside in half of that pair. It is deliberately not an integration, needs no access to your systems, and does not attempt to replace one.