Restaurant Marketing: Fill Tables Without the Delivery Apps

A restaurant doing $40,000 a month in delivery on Uber Eats hands over roughly $12,000 of it in commission, every month, according to Oddle's 2026 breakdown of Australian delivery costs. The same orders taken through the restaurant's own website cost about $2,000. That $10,000-a-month gap is the entire argument for restaurant marketing that owns its channel instead of renting customers from an app.

This guide is about closing that gap. Not by quitting the delivery apps overnight, but by building the channels you control so the apps become one source of customers instead of the only one. Every move below is cheap, most are free, and they compound while a commission payment does not.

The maths on delivery apps

Delivery apps are the most expensive customer a restaurant serves. Uber Eats and DoorDash both charge up to 30% commission on delivery orders on their top tiers in Australia, per Oddle's 2026 analysis. Lower tiers and self-delivery cut that to roughly 15 to 20%, but the discovery-heavy plans that actually bring new customers sit at the top of the range.

Here is what the platforms charge in Australia as of 2026:

Source: Oddle, Own Website vs Uber Eats and DoorDash (2026)

On top of commission come activation fees ($500 for a first Uber Eats location) and sponsored-listing ads that a busy venue can spend $300 to $800 a month on just to stay visible in the app, per the same Oddle data.

Now put that against the margin. The average Australian restaurant nets 5 to 7% profit, and most cafes run at 3 to 6% once rent and award wages are paid, according to Calso's 2026 benchmark. A 30% commission on a business that keeps 6% of revenue does not shave the margin, it erases it. Many venues run delivery orders at a loss and use them to cover fixed costs or chase volume.

The direct comparison, on the same $40,000 monthly delivery volume:

Source: Oddle, 2026. Direct ordering cost covers payment processing (roughly 1.7 to 2.2% plus 30c), ordering software (about 3% all-in), and delivery fulfilment via a service like Uber Direct.

The apps are not worthless. They are a shopfront on a street you do not own, and they reach people who would never find you otherwise. The mistake is treating a discovery channel as your whole business. The 2025 Australian market made that risk concrete when Menulog shut down after 20 years on 26 November, redirecting its customers to Uber Eats, as reported by The Conversation. Every restaurant that depended on Menulog for orders lost that channel with no warning and no customer list to fall back on.

Google Business Profile is your free front door

The single highest-return move in restaurant marketing is a complete, active Google Business Profile, and it costs nothing. It is the panel that shows on Google and Google Maps when someone searches your name or "restaurants near me": your hours, photos, menu, reviews, and a button to call, book, or order.

It matters because that is where diners decide. BrightLocal's 2026 Local Consumer Review Survey found 97% of consumers read reviews for local businesses, and Google's profile is where most of those reviews live. A profile with recent photos, correct hours, and a stream of reviews wins the click before the diner has looked at a single website.

What a restaurant should do with it:

• Fill every field. Cuisine type, price range, hours, dietary options, service areas, and a real menu. Google uses these to match you to searches like "gluten free thai near me".
• Add real photos, often. Owner and customer photos of the food and the room. Profiles with fresh photos get more views and more direction requests.
• Turn on the ordering and reservation links so a diner can act from the profile itself. Point the order link at your own ordering page, not a delivery app, so you keep the margin.
• Post updates. Specials, events, a new menu. Google treats an active profile as a live business and shows it more.
• Reply to every review, good or bad. A calm, specific reply to a complaint reassures the next 100 people who read it more than the complaint worries them.

None of this needs an agency. It needs 30 minutes to set up and 10 minutes a week to keep alive.

Direct online ordering and bookings

Send repeat customers to your own ordering page and you keep the 30% the app would have taken. And customers want you to. In a November 2024 NCR Voyix survey, 58% of customers said they prefer to order delivery through a restaurant's own app or website rather than a third-party platform. The reasons they gave were convenience (65%), easier customisation (50%), and earning loyalty points (36%).

The point is not to ban the apps. It is to use them for what they are good at, first contact, and then move the relationship onto ground you own. A diner who found you on Uber Eats and had a good meal will happily order direct next time if you make it easy and give them a small reason to (free delivery over a threshold, a loyalty stamp, a members-only special).

Practical setup for a venue:

• A prominent "Order Direct" button on your website and Google Business Profile that goes to a commission-free ordering page.
• Online bookings for dine-in, so the phone stops ringing during service. Diners increasingly expect to book in three taps at 11pm, not call at 3pm.
• A flyer or QR code in every delivery bag and on every table that points to the direct ordering page, with an incentive to switch.
• A simple loyalty mechanic. Even a digital stamp card gives customers a reason to come back to you instead of browsing the app.

Direct ordering only works if the page it points to is fast and easy to use, which is the next problem.

A fast mobile site turns lookups into orders

Roughly 70% of restaurant website visits come from a phone, usually from someone hungry and deciding right now. If your site or ordering page is slow, you lose them before the menu loads. Google's research found 53% of mobile visitors abandon a page that takes longer than 3 seconds to load. A slow menu is a customer walking to the restaurant next door.

Most restaurant sites are slow for the same reason most small-business sites are slow. They are built on heavy templates loaded with plugins, sliders, and tracking scripts. Only 43.44% of WordPress sites pass Google's Core Web Vitals on mobile, according to HTTP Archive and Google's CrUX data. A restaurant site that fails those thresholds loads slowly, ranks lower, and converts fewer of the diners who do arrive.

What a restaurant site actually needs is short: the menu (as real text, not a PDF Google cannot read), hours, location with a map, a phone number that dials on tap, an order button, and a booking button. Fast, mobile-first, and honest about what is on offer. Everything else is decoration that slows the page down.

Reviews are the cheapest customers you will ever buy

A steady flow of recent, positive reviews is the highest-return marketing a restaurant can do for free. Harvard Business School economist Michael Luca found that a one-star increase in an independent restaurant's rating raises its revenue by 5 to 9%. The effect held only for independents, not chains, which means a single well-run venue has more to gain from reviews than a franchise ever will.

Reviews also gate whether a diner considers you at all. BrightLocal's 2026 survey found 68% of consumers will only use a business rated four stars or higher, and 47% will not use one with fewer than 20 reviews. Recency matters too: 74% only care about reviews written in the last three months. A venue with a 4.7 rating and reviews stopping 18 months ago reads as a restaurant that has stopped trying.

How to build reviews without buying them:

• Ask, at the right moment. A card on the table or a line on the receipt after a good meal, with a QR code straight to your Google review page. The best time to ask is when the plate is empty and the customer is happy.
• Make it one tap. Every extra step loses people. Link directly to the review form, not your homepage.
• Reply to all of them. Thank the good ones by name, answer the bad ones calmly with a specific fix. Future diners read the replies as closely as the reviews.
• Never fake them. Australian Consumer Law treats fake or incentivised reviews as misleading conduct, and diners spot a wall of five-star reviews posted in one week.

Where to start: the first month

If you own a venue and read this far, here is the order to do it in. None of it needs a marketing budget, only a few hours.

1. Week 1: claim and complete your Google Business Profile. Every field, real photos, correct hours, a real menu. This is the biggest single return and it is free.
2. Week 2: start asking for reviews. Print a QR card that links straight to your Google review page. Put one on every table and in every takeaway bag.
3. Week 3: set up direct ordering and bookings. Stand up a commission-free order page and an online booking link, then add both to your Google profile and website.
4. Week 4: fix the site. Make sure the menu is readable text, the site loads in under 3 seconds on a phone, and the order and book buttons are the first things a hungry visitor sees.

The delivery apps stay in the mix as a discovery channel. Everything you built this month is the machine that turns those one-off app customers into regulars who order direct, at full margin, for years.

Small Business Marketing Ideas That Actually Work (2026)

The channels that bring customers to a small business without a big budget, ranked by return. A companion to this guide for any owner-operator.

Read more

For the wider context on why a fast, owned website beats a rented platform for any small business, our web design service page walks through the performance and ownership case in detail.

Sources

• Oddle: Own Website vs Uber Eats and DoorDash, the Real Cost for Australian Restaurants (2026) - Australian delivery commission rates and direct-ordering cost comparison
• The Conversation / UTS: What a landmark Uber Eats, DoorDash pay deal could mean (Nov 2025) - Menulog market exit and delivery-market change
• Restaurant Dive / NCR Voyix: Most customers prefer ordering delivery direct (Nov 2024) - Direct ordering preference survey
• Harvard Business School: Reviews, Reputation, and Revenue, the Case of Yelp.com (Luca) - One-star rating increase lifts independent restaurant revenue 5 to 9%
• BrightLocal: Local Consumer Review Survey (2026) - Consumer review reading and star-rating thresholds
• Calso: Average Restaurant Profit Margins in Australia (2026) - Australian restaurant and cafe net margins
• Google / SOASTA: Mobile Page Speed Benchmarks - 53% mobile abandonment at 3-plus seconds
• Search Engine Journal / HTTP Archive: 2025 Core Web Vitals CMS Rankings - WordPress mobile Core Web Vitals pass rate