How to get more restaurant orders.

Most restaurants don't have a demand problem. They have a leak problem, and leaks are cheaper to fix than demand is to buy.

Three leaks account for most of it. Every order through a marketplace hands over commonly 25 to 30% before you've bought an ingredient. Every phone that rings unanswered between six and nine is somebody ordering from the restaurant across the road instead. And every checkout that loads slowly, asks for an account, or breaks on a mid-range Android is a full basket abandoned by a hungry person who won't try twice.

Before you spend a euro on ads, measure those three. Count the missed calls for one week. Pull ninety days of platform statements and work out the real cost per order. Then order from your own site on your own phone, on mobile data, and time it. Those three numbers usually point at more money than any campaign will.

Yes, and it's the biggest single change available to most independents. Same food, same kitchen, same driver if you like, but the order comes to you and the whole amount lands in your account.

The arithmetic is easy to check. Take your monthly online sales, apply the commission rate you're actually paying, and compare it with a flat monthly fee plus card processing. Two hundred orders a month at €28 is €5,600 in sales, which is roughly €1,400 to €1,680 of commission at typical rates. Set that against a fixed fee and the answer is usually obvious well before you get to the customer data argument.

Then there's the part that compounds. On your own channel you can see who ordered, what they ordered and when they last did it, which is what makes everything in the rest of this guide possible. The marketplaces will never hand that over. Eclyde runs it at €59 a month plus 5% per order, or €199 a month with no commission, €169 if you pay for the year, and the full comparison is here.

A large share of orders start with a phone and a search: pizza near me, best Indian in Ranelagh, sushi delivery Dublin 4. If your Google Business Profile is thin or out of date, you don't exist at the moment somebody decides.

Claim it and fill it in properly. Every category, exact opening hours including bank holidays, delivery and collection flags, the menu, and an order link that points at your own site rather than a marketplace. Add real photos, taken by you, of your food on your plates in your room. Stock photography of a generic burger reads as fake and it doesn't match what comes out of the kitchen.

Then keep it alive. Reply to every review in a normal voice, especially the bad ones, and never argue in public. Ask for reviews at the moment somebody says the meal was lovely, not by mass email a fortnight later. Recent photos, recent reviews and accurate hours are the parts of local ranking you control, and they cost nothing but ten minutes a week.

By noticing them and getting in touch, which sounds obvious and almost nobody does.

Bringing back somebody who has already eaten your food is far cheaper than finding a stranger, because there's no persuading left to do. The barrier is that most restaurants have no way to reach them. Fix that first: capture an email or a phone number on every order through your own channels, and ask at the counter for dine-in and collection.

Then run two or three simple triggers and leave them alone. A thank you after a first order with a small reason to place a second. A message when a regular hasn't been near you in three or four weeks, sent at the hour they normally order. Something on a birthday. A note when a dish they always order comes back on. Keep any loyalty scheme so simple it fits in one sentence, and make sure the rewards only count on your own channel.

Assume your customer is hungry, tired, holding a child, and on a patchy signal. Every extra tap is an order you don't get.

The target is a minute from opening your site to a confirmed order. That means guest checkout with saved details for anyone who has been before, a one-tap reorder of their last order, Apple Pay and Google Pay on the payment step, and the delivery time and full total shown before anybody starts filling in an address. Photos help, but only real ones of your actual dishes.

Test it yourself, properly. Not on the office wifi on a laptop, but on a mid-range phone on mobile data at seven on a Friday when your own network is busy. If anything makes you sigh, it's costing you baskets.

By moving demand into them, not by discounting the hours that are already full.

A modest offer sent at four on a Tuesday, only to the people who normally order midweek, good for that evening alone, brings forward orders you'd otherwise never see. The same offer sent to everybody on a Friday costs you margin on orders you already had. Targeting is the entire trick, and a standing midweek discount just trains your best customers to wait.

Offers that aren't discounts work even better. A set lunch at a fixed price. A collection-only deal that costs you no driver. Family trays on a Sunday. A pre-order deadline for a big match. Different product at a different time beats the same product cheaper, because it doesn't teach anyone to expect a lower price.

Four, looked at weekly on the same morning, for long enough to see past the weather.

Total orders is the obvious one and the least useful on its own, because it moves with rain and fixtures. Watch the share of orders coming through your own channel instead, because that's the number that decides what you keep. Watch average order value by channel, which tells you whether the attachments are working. Watch how many people ordered a second time within 90 days, which is the honest test of everything in section four. And keep counting missed calls, because that leak reopens quietly whenever the restaurant gets busy.

When something moves, write down one question and go and answer it. Half an hour a week beats a quarterly review that never happens.