Delivery monitoring for a UK restaurant group
Restaurant chains in the UK operate storefronts they do not control on three national platforms, and the gap between what those storefronts show and what head office believes is where the losses sit. Monitoring means reading each listing the way a customer does, per site and per platform, and keeping the result. What makes the UK distinctive is that two platforms have published thresholds you are being scored against, including one that names availability directly, so the measurement is not just useful, it is the one being applied to you.
What has to be measured?
Five things, and only one of them is available from a partner portal in the form you need.
Whether the listing was orderable during your published hours. Whether the prices shown match the brand standard. Whether configured promotions are visible to a customer. How far the delivery zone actually reaches from real addresses. And the customer rating and its direction.
Portals give you the last one and give you settings rather than outcomes for the rest.
Why is availability the first priority?
Because nothing else matters while a site is off, and because the UK platforms grade it explicitly.
Just Eat’s Local Legend criteria include “Time offline at 10% or below”, held “for two consecutive quarters”, alongside order volume, review rating and an “FSA rating of 3+ or a ‘Pass'”. Deliveroo does not publish a percentage but reserves the right to withdraw programme access “at our sole option if at any time you do not meet our eligibility requirements”.
So availability is simultaneously the largest direct revenue effect and an input to commercial programmes that are worth more than the individual outages.
What does the UK add that other markets do not?
Hygiene ratings as a live commercial variable, on a public register, with a slow clock.
Deliveroo requires “at least 2 to list on the Deliveroo platform” and three for Marketer access. Just Eat’s partner guidance sets “a minimum of 3* or Pass in Scotland … to sign-up and remain trading”. Those values change weeks after an inspection, without notice to anybody in the business, and they gate promotional access.
That makes rating publication dates part of a monitoring programme rather than a compliance footnote, and it is the single most UK-specific thing on the list.
Why not just watch the sales?
Sales cannot distinguish a quiet evening from an invisible storefront. One row in the report covers both.
Working backwards from revenue gives you suspicious periods rather than incidents, and suspicion does not survive a conversation with a site or an account manager. Availability has to be observed directly, from the customer side, at an interval short enough that an evening outage cannot hide between checks.
Kitchain (kitchain.co) reads listings on a schedule and keeps them, and a kept series is the whole difference between ranking your sites on availability and having an opinion about them.
How should a group start?
With one month of measurement and no changes, then a ranked list.
Resist the urge to act during the first month. What you want at the end of it is the distribution: which site and platform combinations carry the lost hours, and when in the day the losses fall. Whatever shape that comes out in is the shape that decides where the effort goes, and no amount of reasoning in advance produces it.
Acting before you have the distribution spreads effort evenly across sites that mostly do not need it.
What should it cost in attention once running?
Ten minutes each morning and one short exception list a week, held by one named person.
Anything heavier gets abandoned. The value is not in the volume of data but in somebody seeing three lines before service and making three phone calls.