Running Your Kitchen: From Order In to Plate Out

Kitchens don't break all at once. They slip one ticket at a time, until a table has been waiting twenty minutes, a driver is parked at the door, and the cashier is apologising for something they can't even see. Restaurant kitchen management isn't a chef's gift, it's a system: every item knows its station, every ticket knows its clock, and every delay turns into a number you can see before an unhappy guest hands you one. This guide walks the whole path with you, from the second an order lands to the second the plate leaves the pass: how to split stations, how to route items, how to order the queue, how to make a table's plates land together, and how to read timing numbers instead of ruling by gut.

Let's agree on one definition first. A kitchen that works isn't one that cooks fast, it's one that pushes tickets out in a time you can predict. That gap is huge. Take a hypothetical case to make the point: a kitchen that plates in six minutes when it's quiet and twenty-five at peak. That swing is what wrecks your service and makes your reviews bounce between excellent and awful in the same week. An organised kitchen might not be the fastest, but its time stays close to itself on most tickets, and more importantly it knows it's about to run late before it runs late. Everything below this line serves two goals only: hold that time steady, and give you an early warning the moment it starts stretching.

Before You Buy a Screen: Four Decisions You Make on Paper

A screen doesn't fix a disorganised kitchen, it just exposes one. Whatever system you install will ask you the same four questions, and your answers shape your service for years: what your stations are, where each item goes, how long each item takes, and which doors orders come through. Sit down with your chef for half an hour and sketch it on paper before you open a single setting. Walk into any POS without those answers and you'll configure it twice, train your team twice, and spend the second round explaining why you changed what you told them last week.

1. Stations: Split by Flow, Not by Equipment

One mistake that keeps showing up is splitting stations by equipment: oven station, fryer station, grill station. That's a useful map for maintenance, not for service. The right split follows flow: which items move through the same hands in the same window of time. If the person on the grill is also the person shaking the fryer basket, you don't have two stations, you have one station with two names, and the first busy night will tell you so. Stand in the kitchen for a full peak hour and watch the hands instead of the machines, and your real stations will show themselves.

Here's a working rule that ends a long argument fast. If one of two stations always finishes well ahead of the other, consistently, you haven't split the work, you've built waiting into the design. Either merge them, or move an item or two off the loaded station onto the idle one. The reverse matters just as much: if one station is always the last to finish on every ticket, it sets the pace of your whole kitchen no matter how quick everyone else is. A kitchen runs at the speed of its slowest station, never at the average of all of them.

And the "how many screens do I need" question follows the same logic: a screen serves a position where a pair of hands is working, not a piece of equipment. A whole small kitchen can run on one screen showing every station as its own zone, while a kitchen with three spread-out positions needs three display points so nobody walks five paces every time they want to read a ticket. Working rule: any station whose cook has to leave their spot to see the order is a station without a screen, even if there's one hanging two metres away.

2. Routing: Every Item Knows Where It Goes, and Combos Get Written Twice

Routing is the rule that says this item appears on that station. Sounds obvious, and it is, until you hit composed items. A burger with fries isn't one item on one station, it's two items on two stations, and both have to appear at the same instant or you end up with one station working and another that doesn't know work exists. Modifiers behave the same way: extra cheese belongs to the grill, a side sauce belongs to cold prep. If a modifier lands on the wrong station, an incomplete plate leaves the pass and comes straight back.

And there's a third column that usually drops out of the plan: not just where an item goes, but when its station should start. A three-minute item shouldn't start alongside a nine-minute item, because it'll spend six minutes sitting there going cold. That's the start offset, and it's the difference between a table eating together and a table where half the food is lukewarm. Write the table below for your ten best sellers and pin it up in the kitchen for the first week:

3. Standard Time: The Number Everything Else Hangs On

Standard time is how long an item takes from the moment the station starts to the moment it's ready to serve. Without that number, any conversation about lateness is a feeling rather than a measurement, and every dispute between floor and kitchen ends with whoever talks loudest. Get it yourself, don't take it from a supplier or a template: grab a watch, stand over your ten best sellers, and time each one during real service. Don't measure at three in the afternoon with an empty kitchen, because that number will lie to you at precisely the moment you need it.

Time each item three times and take the highest, not the average, because you're building a promise to a guest rather than a number to be proud of. Re-measure whenever something real changes: a new recipe, new equipment, a new crew. And the point of the number isn't display, it's control. Any ticket that runs clearly past its standard time becomes a case worth asking about the same night, not at month end when nobody remembers what happened that evening.

4. Channels: Which Doors Orders Come Through

You have one kitchen, but more than one door knocking on it: the dining room, delivery apps, counter takeaway, your own online store, and the QR code on the table. Each door carries a different expectation about time, and that's where the mess starts. Push every door into the same queue with the same logic and you'll find a delivery order scheduled for an hour from now occupying a station right this second, while a table sitting in front of you waits behind it. Separate the logic from day one:

• Dine-in: starts the moment the order is saved, and the guest is sitting there, so lateness turns into a complaint at the table.
• Counter takeaway: almost the same logic as dine-in, minus the coursing, since it all gets bagged in one go.
• Delivery apps: you don't control when the driver arrives, so you start so the food finishes just before they do, not the second the order lands.
• Your online store and QR ordering: you set the pickup or serving time, so you can schedule when the kitchen sees it instead of taking it all in one burst.
• Scheduled and event orders: they hit the kitchen at execution time and get deducted from that hour's capacity before you accept anything else onto it.

The point is that delivery and dine-in don't share urgency fairly: a delivery ticket is waiting on a driver, a table ticket is waiting on a human who can see you. When you pull delivery apps into one screen alongside your own store and QR orders, you get a single source for the queue and one logic to order it by, instead of a crew hopping between devices. And if you want the other half of the picture, what each channel does to your margin rather than your clock, working out delivery app commissions fills that in.

Order In to Plate Out: Seven Steps You Set Up Once

None of what follows needs budget. It needs a decision and one disciplined week. Build them in order, because each one leans on the last: you can't order a queue without standard times, and you can't measure lateness without a clear hand-off point. And one piece of advice from how change actually lands in a working kitchen: don't roll all of it out in a single night. Take a step or two a week, and apply it to your ten best sellers only at first, because your traffic usually concentrates there and your own sales report will tell you whether it does, and the rest of the menu will fall in behind them once the rule settles. A kitchen accepts change when it sees the result the same week, and resists it when it arrives all at once.

1. 1
   Price the time before you price the plate

   Get a standard time for your ten best sellers during real service and post it where everyone can see it. That number becomes your reference in every floor-versus-kitchen argument, and the base for any delay alert you build later.

2. 2
   Map items to stations and lock the routing

   Every menu item gets one known station, and composed items open on two stations at the same instant. Loqma's kitchen screen sends each line to its station the moment the order is saved, so nothing sits unclaimed waiting for someone to shout it across the pass.

3. 3
   Pick one rule for ordering the queue

   Choose a single logic and stick to it: oldest first, or nearest to its due time first. Mixing the two is exactly how one ticket sits for half an hour while later ones sail out. A clear rule beats a perfect rule.

4. 4
   Tie a table's items together with start offsets

   Work out when each item should start so it finishes with the slowest item on the same ticket. The three-minute item starts later than the nine-minute one, so every plate is ready in the same moment instead of one going cold waiting on another.

5. 5
   Split the logic for channels and scheduled orders

   Dine-in starts on save, delivery starts at pickup time minus prep time, and a scheduled order comes out of its own hour's capacity rather than this minute's. That's how an order due in two hours stops elbowing a table sitting in front of you.

6. 6
   Mark the ready point and the hand-off point

   The station marks the line ready, and the pass marks the ticket handed off. Those two marks are the source of every number later: without them you have tickets that go in and vanish, with them you have a real time per ticket and per station.

7. 7
   Read the numbers weekly and change one thing

   Spend half an hour a week on ticket times and pick exactly one change: an item moves station, a standard time gets updated, or a dish comes off the peak menu. One change a week compounds; ten at once just confuses your crew.

Hunt the One Bottleneck, the Rest Is Noise

When a night gets pressured, go looking for the single station that's causing it, and you'll usually find everything else is a symptom trailing behind it. The grill runs late, so a server stands waiting on it, so another table slips, and you walk away with the impression that everything is slow. Try to treat all the symptoms at once and you'll burn the effort for nothing: lots of changes, no felt difference. What works is spotting the station whose tickets stack up first under pressure, fixing that one alone, and watching the rest of the line relax on its own.

And fixing a bottleneck rarely means hiring. Try it in this order: move an item or two onto a lighter station, prep the longest-lead component an hour before the rush, simplify the slowest dish or pull it from peak hours entirely, and if it's still tight after all that, then yes, you genuinely need a second pair of hands on that station for two specific hours rather than the whole shift. That sequence saves you entire hiring decisions, and more importantly it shows you exactly where the problem is before you pay for it.

Coursing: The Table Eats Together or It Doesn't Eat

A lot of the complaints written as "the food was cold" are really a coursing problem, not a heat problem. One plate was up at minute four and sat waiting for its partner at minute nine, so they reached the table together but one of them had already lived five minutes on a shelf. The fix isn't speeding up the slow one, it's deliberately holding back the fast one. That idea needs explaining to the crew the first time, because it reads as backwards: why are we slowing down when we're trying to hurry.

Take one four-item ticket and do the arithmetic in front of the team once and they'll have it. The slowest item sets the ticket's clock, and every other item starts later by the difference between its own time and the slowest one's. A nine-minute burger starts immediately, four-minute fries start after five, a three-minute salad starts after six, and a two-minute drink starts after seven. All four are done at minute nine, and the whole ticket leaves the pass at once instead of in two waves with a server walking back and forth.

There's one exception you need to teach the crew on day one, or coursing will turn on you: not every table wants its food together. A table that ordered starters before mains wants a deliberate gap, and a rushed guest at lunch wants whatever is ready right now. So coursing is a default, not a law, and the floor needs a way to tell the kitchen "this table is eating in courses" or "this one is in a hurry". Without that signal you end up with a disciplined kitchen delivering service that suits half your guests, which is worse than chaos because it's tidy chaos.

Peak and Scheduled Orders: Same Kitchen, Different Logic

Peak isn't a longer hour, it's an hour where the queue fills faster than it empties. Which means every ticket that lands at the start of the rush is paid for by every ticket behind it. So the decisions you make before the rush matter far more than the ones you make during it: which items you prep ahead, which items you pull from the menu for those two hours, and how many delivery orders you'll run in parallel before you pause intake. Pausing intake for ten minutes at the right moment is far kinder to your reputation than accepting ten orders that all go out late.

Scheduled orders need the logic inverted: you know the hand-off time in advance, so you count backwards from the end. An order collected at nine that takes fifteen minutes to make hits the kitchen at eight forty-five, not at seven when it was booked. Bakery, pastry and event orders are a special case here, because their prep is measured in hours rather than minutes, so they come out of a whole day's capacity and get reserved in advance instead of being accepted and discovered on the morning. Any scheduled order you accept without deducting it from that hour's capacity is a promise you don't know you can keep.

When an Item Runs Out Mid-Service

The most expensive ticket you'll take isn't the late one, it's the one you accept and can't fulfil. A guest who waits twelve minutes and then hears the item is finished is a guest you lost twice: once on time and once on trust. And the trouble is that a sold-out item reaches the till fast but doesn't reach the delivery apps, the online store or the QR menu, so orders keep coming in through doors you'd forgotten were open.

The structural fix is to tie stock to the recipe, so the ingredient is deducted on every sale and the item takes itself off every channel once it hits the level you set. That lifts the remembering off your crew at the exact moment they need to be focused elsewhere. And whether or not your system does that automatically, keep a written rule: whoever discovers the shortage kills the item on all channels first, and only then goes looking for a substitute. That order of operations isn't a detail, it's the difference between a small adjustment and a wave of apologies.

Edits and Cancellations Mid-Service: Who Can Do What

Every kitchen has tickets that change after they've landed: a guest changed their mind, a line was rung wrong, a table split. That's normal and nobody should be upset about it. What needs structure is the question of who can change it, when, and why. If anyone can pull a line off a ticket that's already live on a station and leave no trace, you haven't just lost a sale, you've lost the ability to understand your own evening: the ticket that vanished off the screen shows up in no number later, and the station that already made the item burned that effort with no explanation.

The right control is simple and it isn't aimed at anyone: permissions follow the role rather than personal trust, a cancellation or a removal asks for a reason picked from a short list, and the sensitive actions go through a manager approval with a PIN. And here's the part to say to your crew plainly: that log isn't an accusation tool, it's an explanation tool. A number telling you removals cluster in one particular half hour on one particular station usually means you have a design problem in that half hour, not that anyone is doing anything. The system shows you the pattern; you're the one who asks and finds the cause.

Shift Handover: The Kitchen Doesn't Start From Zero Every Night

Treat every shift like a brand new day and you waste the five most useful minutes of it. The shift that's ending knows things the next one needs: which item ran out and came back, which station was struggling, which tickets came back and why, and what was prepped ahead and how much of it is left. Five written minutes at the pass carry that across instead of letting it evaporate when people clock out.

Tie it to something formal so it doesn't rely on enthusiasm: the kitchen handover happens at the same moment the cashier closes their register shift and pulls the closing report, so you get one timestamp separating two sets of responsibility. Then if a stock gap or a stuck ticket turns up, you know which shift it belongs to instead of combing through a whole day. That's the same idea that makes any system's reports worth reading in the first place: every number needs a time window and a person attached to it, or it's just a remark nobody acts on.

What Changes After a Month: The Numbers You Read

After a month of running this way you'll have something you didn't have before: history. Every ticket has a time in and a time out, every station has a record, every evening has a shape. That's where real management starts, because you stop asking "why is service slow" and start asking "which station, which hour, which item". That shift is the whole thing; the rest is detail.

The first number to get used to is the median ticket time. The second is the one that matters more and gets forgotten most: the delay tail, the slice that came out far slower than the rest. The median reassures you; the tail writes your reviews. Take this hypothetical hundred tickets from a single evening and watch two numbers tell two opposite stories about the same night:

Notice the trap. Look at the median alone and your kitchen is excellent. But twenty-five tickets ran past fourteen minutes, and those are exactly the people who write about you. So the right question isn't "what's my average ticket time", it's "what do the slow tickets have in common". Usually one answer repeats: the same item, the same station, or the same half hour of the evening. Any one of those three you can fix with a simple operational decision, without adding a person or a machine.

• Median time per station: shows you the station dragging your kitchen back every single day, not once a week.
• The delay tail: how many tickets blew past their standard time, and which hour they clustered in.
• Slowest items: any dish showing up in the slow tickets more than its share of sales would explain.
• Tickets sent back to the kitchen: a missing or wrong line, which is a routing signal, not a skill signal.
• Channel mix during peak: exactly which channel is crowding the dining room.

One last thing: don't compare an evening to an evening. Compare Thursday to Thursday and Tuesday to Tuesday, because your week isn't flat and you already know that from your own till. Holding Thursday against Tuesday will show you a decline in performance that isn't real, it's just a different level of pressure. Over time you'll find every day has a personality: one with lots of small tickets, one with fewer but heavier ones landing on a single station, one where delivery crowds out the dining room. Once you know each day's character, you have a real basis for scheduling kitchen shifts and prep instead of spreading them evenly and then complaining about Monday and Thursday in the same tone of voice.

None of these need a spreadsheet or a person keying them in, as long as the kitchen screen and the point of sale are the same system: the order is saved once, routed to its station, and lands in the reports with its own clock, no second entry. If you're comparing systems right now, make reporting part of the comparison rather than an afterthought, and look at the plans and what's included with kitchen, stock and reports counted inside the package instead of as extras that surface later. And speaking of what surfaces later, food safety and in-kitchen record-keeping requirements vary by activity and get updated over time, so check the Saudi Food and Drug Authority before you lock in an internal procedure and build your records on it.