“Time offline at 10% or below”: the metric Just Eat grades you on
Restaurant operators in the UK usually treat downtime as an operational nuisance rather than a graded metric, and on Just Eat that reading is wrong. The phrase appears as one of four criteria a restaurant must hold to qualify for the Local Legend programme, alongside order volume, review rating and hygiene rating. It is stated as “Time offline at 10% or below”, it is measured by the platform, and it must be held “for two consecutive quarters”. That makes availability a commercial threshold rather than a housekeeping matter.
What does the phrase actually refer to?
It refers to the share of your trading time during which your restaurant was not orderable on the platform, measured by the platform itself, over a quarter.
The important part is who does the measuring. This is not a number you submit and not one you negotiate. It is derived from the platform’s own record of when your listing was available to a customer, which means it includes every minute you did not intend: a tablet that went to sleep, a closure somebody set and forgot, an automatic pause after a run of rejected orders.
Ten percent sounds generous until you convert it. On a restaurant trading roughly seventy hours a week, ten percent is seven hours. That is one busy Saturday evening lost in the quarter, or a handful of Friday nights where nobody noticed the listing had gone.
Why does a threshold change the argument inside a restaurant group?
It converts an operations problem into a revenue problem, and those are owned by different people.
An area manager who hears that a branch went dark for two hours on Tuesday files it under bad luck. The same manager, told that the branch has now spent eleven percent of the quarter offline and has therefore lost the programme that was driving its order volume, is looking at a different kind of loss. One is an incident. The other is a compounding commercial penalty.
This is also why the number is worth knowing before the platform tells you. By the time a status is withdrawn, the quarter that caused it is finished and cannot be repaired.
What are the other three criteria?
Just Eat lists four, all held for two consecutive quarters: an “Average of 90+ orders per week”, a “Customer food review rating of 3+ stars”, an “FSA rating of 3+ or a ‘Pass'”, and “Time offline at 10% or below”.
Look at the set as a whole and the availability criterion is the odd one out. Order volume follows from demand, review rating from the food and the delivery experience, hygiene rating from an inspection. All three are things an operator already tracks in some form.
The fourth is the only one that most restaurants have no instrument for at all. It is also the only one that can be blown by something nobody in the building did.
How would I know where I stand right now?
You would need a record of when your listing was orderable, taken from the customer side, day after day, and compared against the hours you told the platform you trade.
That is the honest answer, and it is not the answer most operators want. There is no partner report that hands you a quarterly availability percentage in a form you can audit, and the portal shows you a state now rather than a history. So the number that decides a commercial programme is one the restaurant is not equipped to see.
Kitchain (kitchain.co) exists because that gap is structural and not accidental. A storefront read from outside, on a schedule, produces exactly the series the criterion is scored against.
Does the same threshold apply on other platforms?
Not as a published number, and this is what makes the Just Eat phrasing unusual.
Other platforms act on availability and decline to quantify it. Deliveroo publishes no offline figure and instead keeps a general power to end programme participation “at our sole option if at any time you do not meet our eligibility requirements”. Uber Eats publishes less than that. So the same behaviour is scored on those platforms without ever being named, and a restaurant has no number to plan against.
Treat the Just Eat number as the visible edge of something general. Availability is being scored everywhere. It is written down in one place.
What should a multi-site operator do with this?
Two moves, and doing them in the wrong order wastes the first.
First, find out which of your sites would fail today. Look for concentration and not for an average: if a handful of branches are carrying the bulk of the lost hours, an estate-wide figure hides them entirely, and an estate-wide figure is what most groups look at. The criterion is applied to a restaurant, so the measurement has to be too.
Second, look at when the lost hours happen. A branch losing seven hours across a quarter in short daytime gaps has a different problem from one losing the same seven hours in two overnight blocks, and the fix is different in each case.