Restaurant marketing ideas that work.

With the people who've already eaten your food, and with the search results in your own postcode. Both are free, both work in weeks rather than quarters, and neither needs an agency.

Everything else on this page is downstream of two assets: a list of customers you can contact, and a listing that shows up when somebody nearby is hungry. Build those first. Paid ads pointed at a restaurant with no follow-up and a half-finished Google profile are money set on fire.

The other reason to start there is time. You have about forty spare minutes a week, realistically, and most of them are on a Tuesday morning. Pick tactics that survive that constraint.

By noticing they've gone, which is the part almost nobody does.

Work out your normal gap. If your regulars order every ten days or so, somebody at four weeks isn't busy, they've drifted, and there's usually a reason: a bad delivery, a dish they didn't like, or a new place that opened on the corner. A short message at that moment does more than any campaign. Keep it plain, keep it human, and put a small reason to act in it. A free side beats a percentage off, because a percentage reads like a supermarket and a side reads like a gesture.

A few practical points. Send it at the hour those people usually order, not at nine in the morning. Cap it so nobody gets chased twice in a month. And make sure the offer only works on your own channel, otherwise you're paying commission to win back your own customer.

This is dull, repetitive work with real money in it, which is exactly what marketing automation is for.

A boring one. Stamp cards get left in the car. Points with tiers, expiry dates and a separate app to download get abandoned at the download.

The version that works is the one people understand in a sentence and never have to carry: order ten times, the eleventh is on us, tracked automatically against the phone number or email they already give you. Set the threshold where it's genuinely reachable for a regular within a couple of months, then say it out loud at the counter and print the progress on the receipt.

Most 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, you don't exist for that moment.

Claim it, then fill in every field properly. Categories, exact opening hours including bank holidays, whether you do delivery and collection, the menu, and a link that goes to your own ordering page rather than a marketplace. Photos matter more than owners expect, and they must be your food, shot by you, on your plates, in your room. Stock photography of a generic burger reads as fake to customers and it doesn't match what arrives at the door.

Then keep it moving. Reply to every review, good and bad, in a normal voice, and never argue. Ask for reviews at the right moment, which is when somebody says the meal was lovely on the way out, not in a mass email a fortnight later. Post occasionally: a special, a change in hours, something new on. Recent activity, recent photos and a steady drip of reviews are the parts of local ranking you actually control.

With something physical in the bag, and a reason that's worth the change of habit.

Every marketplace order is a delivered advert with your food inside it. Put a card in the bag with a short line, a QR code that opens your ordering page, and one clear incentive for ordering direct next time. Print it on decent card so it doesn't come out greasy. A sticker or a printed sleeve on the box works too, because it survives the journey better than a loose flyer.

The incentive has to be real. People aren't switching apps to save twenty cent. A free side on the first direct order, or loyalty stamps that only count on your own site, is enough to move a decent share. Then the follow-up matters more than the card: once they've ordered direct once, you have their details and the rest is your own list.

Commission on the apps commonly runs 25 to 30% per order, so every regular you move across is worth €7.50 to €9 on a €30 order, every time they order, forever.

Your own food, your own people, and what's actually happening this week. Posting three times a day is not the goal, and it never survives a busy fortnight anyway.

A short video of a pizza coming out of the oven, shot on a phone, beats a designed graphic every time. So does the specials board, a new starter, the kitchen at seven on a Friday, or the person who's worked with you for nine years. Show the room as it looks, not as a stock library imagines it. If a picture doesn't look like what a customer will get, it costs you trust when the box opens.

Two per week, consistently, is worth more than a burst of daily posts followed by six silent weeks. And put the link to your own ordering page in the profile, not a marketplace link, because you're paying for that traffic with your own time.

Relevance and restraint. The two failures are sending the same message to everybody, and sending too often.

Use what you already know. Somebody who has ordered a vegetarian main four times running shouldn't get a message about the new steak. Somebody who orders every Friday shouldn't get a Friday offer, they were coming anyway, and you've just given away margin. Segment on two things to begin with: what they order and when they last ordered. That's enough for most restaurants.

Keep SMS short and rare. It costs money per message and it's read within minutes, so save it for the things that deserve interrupting somebody: a genuine offer with a deadline, or a heads-up that the restaurant is closed on Monday. Email carries more detail and costs nothing, so it can go out more often.

One legal point worth getting right in Ireland and the UK: you need consent to market by email or SMS, every message needs an easy way out, and existing customers you're contacting about similar food are the narrow exception rather than a free pass. Keep a record of when and how somebody opted in, and honour an unsubscribe immediately.

Make it small, make it targeted, and give it an end date. A standing discount is not marketing, it's a price cut with a poster.

The test to apply before you send anything: would these people have ordered anyway? A 20% off code on a Friday to your best customers costs you money and buys you nothing. The same offer at four on a wet Tuesday, sent only to people who normally order midweek, good for that evening only, brings forward demand you were never going to get.

Prefer offers that cost you food cost rather than cash. A free garlic bread costs you maybe eighty cent and reads as a couple of euro. Free delivery on a collection-heavy night costs you nothing at all if your radius is short. And cap the length: two hours or one evening, so it stays a reason to act rather than a new price list.

After the free work, not before, and with a number you're willing to lose.

When you do start, keep the radius tight. A restaurant's realistic catchment is a few kilometres, and every euro shown to somebody outside it is wasted. Start with a small daily budget, run it for a full fortnight rather than a weekend, and send the click to a page where somebody can order in under a minute. Most restaurant ad spend is lost after the click rather than before it.

Measure it honestly. Not impressions, not likes, not reach. Orders attributable to the campaign, minus the food cost, minus the spend. If you can't work that out, you're not ready to spend more. And judge every euro of ad spend against the alternative: the same money put into a bag insert or a win-back message aimed at somebody who already likes your food.

Pick one number per tactic before you start, and write down what it is today.

For local search, it's how many people found you and rang or clicked through from Google in a month, which Google's own profile insights will show you. For win-back messages, it's how many of the people you contacted ordered within a fortnight. For loyalty, it's the share of orders from repeat customers over 90 days. For bag inserts, it's the number of first-time direct orders each month.

Give each one at least a month, ideally a quarter, before you judge it. Restaurant weeks are noisy: a wet fortnight, a bank holiday or a local match will move your numbers more than most campaigns. What you're looking for is a trend that survives the weather.