Paid ads
Facebook & Instagram ads for restaurants
Run right, restaurant Facebook ads are the fastest way to fill a service you choose. We build tight, geo-targeted Meta campaigns that put your room in front of the few thousand locals who can actually drive to you, and we judge every dollar on cost per booked cover, not clicks or likes. It is part of Packed Out.
Most venue ad spend is just boosted posts.
Here is how it usually goes. The post is doing alright, so the owner hits the blue Boost button, throws fifty dollars at it, and Facebook reports a few thousand "people reached". Feels productive. Then the week ends and nobody can point to a single booking that came from it. The money went to likes from people who will never set foot in the place.
Boosting is the trap because it is easy and it shows you a big number. But it skips everything that makes ads work: a real objective, a tight location, the right audience, a tracked action, and creative built to test. Without those, you are paying Meta to show your post to whoever is cheapest to reach, which is rarely a hungry local near your door.
The flip side is that Meta is one of the few places you can fill a specific quiet Tuesday this week. The targeting and the speed are genuinely powerful. They just have to be pointed at a booking, not a vanity metric.
We don't buy reach. We buy covers, and we can prove which campaign, which audience and which ad delivered them.
Campaigns built to fill the room.
Strategy & offer
We start with what you need filled: a quiet midweek service, a new menu, a function push. Then we build the offer and the campaign around it, because an ad with no reason to come in is just noise.
Tight geo & audience targeting
A radius around your venue, plus the suburbs your diners actually come from. Lookalikes off your customer list and past bookings, and retargeting of people who already engaged. No budget sprayed across a whole city.
Creative direction & testing
Ad creative that looks like your venue, scripted and shaped from your footage or ours, then split-tested. We run several angles, keep the winners and refresh them before they tire.
Tracking & pixel setup
Meta pixel and conversions API installed so we can see real outcomes: bookings, offer claims, calls. Without tracking you are flying blind, so we get this right before we spend a dollar.
Weekly optimisation
Campaigns watched and adjusted weekly, not set and forgotten. We shift budget to what is converting, cut what is not, and manage frequency so you are not burning the same people out.
Reporting in cost per cover
Each month, the one number that matters: what a booked cover cost you, plus spend, return and what we changed. No screenshots of reach dressed up as success.
Track first, then scale what pays.
Set up tracking
Pixel, conversions API and a clear conversion action wired up before launch. We need to see bookings, not guess at them, so this comes first every time.
Build the offer & creative
We pick the service to fill, craft an offer that protects your margin, and produce a handful of ad variations to test against each other.
Launch tight & test
Campaigns go live to a focused local audience. We let the data come in, then read which audience, placement and creative is actually producing bookings.
Scale & report
Budget moves to the winners, losers get cut, frequency stays healthy. Each month you see cost per cover and return, in plain numbers.
Spend you can actually account for.
Every dollar tied to a result you can see in the booking system.
- The ability to fill a specific quiet service this week, on demand.
- Ads shown to nearby locals who can realistically drive to you, not a whole city.
- A tracked path from ad to booking, so you know what worked.
- A clear cost per booked cover, the number that tells you if it is worth it.
- You own the ad account, the pixel and the audience data. Always.
- Boosted posts that buy likes and nothing else.
- A budget sprayed across people who will never visit.
- Reports of "reach" with no link to a single booking.
Targeting that fits how we're spread out.
Australian cities are spread out, so radius targeting matters more here than in dense overseas markets. A diner will happily cross a few suburbs in Brisbane or Perth but not the whole metro, so we set the radius to the real catchment your venue draws from, and we lean harder on it for venues outside the tight inner-city grids.
Timing is everything for venue ads. We push offers ahead of the moments that fill rooms here: long weekends, Melbourne Cup, footy finals, end-of-year functions from late October, and Valentine's and Mother's Day for the venues they suit. We also use ads to defend the quiet stretches, like the long February lull, when a well-placed midweek offer earns its keep.
And we keep the spend honest. Any ad involving alcohol follows Meta's restrictions and Australian responsible-service rules, and we keep your ad account and customer data in your own hands so nothing is locked to us. You can see exactly what went to Meta and what went to running the work.
Want the mechanics? Module 10 of the Playbook covers paid ads for venues end to end, from audiences to creative to reading the numbers.
Get the plumbing right before you spend a cent.
Most venue ad accounts are a mess of half-finished boosts with no structure underneath them. That is why the spend never adds up. Before we run anything, we set the account up the way it should have been from the start.
That means a Business Manager that you own, with your page, your ad account, your pixel and your payment method all sitting under it. Not ours. If we part ways, you keep the lot, including the history and the audiences you have built. We have seen too many owners locked out of their own data because an agency built everything inside their own account. We do the opposite.
Inside the account, campaigns are kept clean and separate by job. One campaign chasing awareness should never share a budget with one chasing bookings, because Meta will quietly feed the cheaper-to-serve one and starve the other. We keep budgets, audiences and objectives apart so each can be read on its own and turned up or down without dragging the rest with it. We also set conversion events properly in Events Manager and rank them, so Meta knows a booked table is worth more than a page view. Skip that and the system optimises toward whatever is easy, which is almost never a cover.
The three campaigns every venue needs.
Not ten clever ideas. Three jobs, run properly, feeding each other.
Awareness, to the right locals
A top-of-funnel campaign that shows your best food and room footage to people inside your catchment who have never heard of you. Cheap to run, judged on reach into the right postcodes and on how many of those people we can later retarget. This is the campaign that fills the pool the other two fish from.
Offer and conversion
The campaign that asks for the booking. A specific reason to come in, pointed at the service you want filled, optimised for the booking or offer claim, not the click. This is where the budget weight goes, because this is the one that produces covers you can count.
Retargeting
The cheapest covers you will ever buy. People who watched your video, visited the booking page or engaged with the page get a second, sharper nudge. They already know you. They just needed reminding. We cap frequency so it works as a tap on the shoulder, not a nag.
These three run together. Awareness feeds the retargeting audience, conversion turns warm people into bookings, and we watch how budget flows between them every week.
The pixel, the API, and why both matter now.
The Meta pixel is a small piece of code on your site that tells Meta what people do after they click: view the menu, start a booking, finish one. That signal is what lets the system go and find more people like the ones who actually booked. Without it, you are guessing.
The problem is that browser tracking has been quietly breaking for years. Apple's privacy changes, ad blockers and cookie limits mean the pixel alone now misses a big share of what really happens. That is where the Conversions API comes in. It sends the same events to Meta server to server, from your booking system or website backend, so a booking still gets counted even when the browser signal is lost. Running the pixel and the Conversions API together is the difference between Meta seeing most of your results and seeing half of them.
We wire both up before launch, test that a real booking fires a real event, and make sure the data deduplicates so one booking is not counted twice. Then we set the conversion as the thing campaigns optimise toward. Get this layer right and everything downstream, the targeting, the scaling, the reporting, actually works. Get it wrong and you are flying blind no matter how good the ads look.
If an agency wants to spend your money before the pixel and Conversions API are firing on a test booking, that is the moment to stop.
Three audiences that actually convert.
Geo-radius
A drop-pin radius around the venue, tightened or widened to match how far people really travel to you. For a suburban café that might be three kilometres. For a destination bar, fifteen. We layer suburbs and postcodes on top so the spend lands on people who can realistically be at your door tonight.
Lookalikes
We upload your customer list and past bookings, then ask Meta to find more people who resemble them inside your catchment. Lookalikes built off real buyers beat interest targeting almost every time, because they are modelled on people who have already paid you, not people who once liked a food page.
Custom audiences
Your warm pool: site visitors, video viewers, people who engaged with the page or Instagram, and your own contact list for retargeting. These are the people closest to booking, so they get the offer and the reminder, not the cold introduction.
Your own footage beats a polished ad.
The creative that works for venues is almost never the glossy, over-produced kind. It is a real phone video of the pasta being finished, the room filling up on a Friday, the bartender making the drink people come back for. It looks like the feed it sits in, so people watch it instead of scrolling past, and it looks like you, so the diner who turns up gets what they were sold.
We shape this from your own content. A short list of the shots that sell, a simple way to film them during service, and then we cut, caption and script them into ads. Captions matter because most of this is watched on mute. The first second matters most, because that is all you get before the thumb moves.
Then we test. Three or four angles against each other: the dish, the room, the offer, the regular's review. We keep what produces bookings, kill what does not, and refresh the winners before they tire, because the same ad shown too long to the same people stops working. This is ongoing, not a one-off shoot.
Cost per booked cover, not cost per click.
Here is the number that matters: what did one booked cover cost you. Take the spend, divide by the covers the campaign actually produced. If a diner spends sixty dollars with you and a cover costs you eight dollars to acquire, that is a campaign worth scaling. CPC, CPM, reach and engagement are diagnostics at best. They do not pay the rent.
Most agency reports hide behind the soft numbers because they look good and prove nothing. A million impressions and a six-cent click mean nothing if the till did not move. We tie spend to bookings through the tracking, then report the cost per cover and the return, in plain dollars, every month. If the maths does not work, we tell you and we change it.
We also start small on purpose. A test budget long enough for Meta to learn and for us to see what a cover really costs, before a dollar more goes in. A few hundred to around a thousand a month fills specific quiet services for most independents, because the radius is tight. The budget follows the result, never the other way around.
If you cannot point at the covers a campaign produced, you are not running ads. You are donating to Meta.
When it works, push it. When it doesn't, stop.
Boosting vs real ads
Boosting buys reach to whoever is cheapest. A real campaign buys a tracked outcome from a chosen audience. One leaves you with likes you cannot bank. The other leaves you with covers and a record of which ad produced them. We only run the second kind.
Scaling without breaking it
Once a campaign is producing covers under your target cost, we scale it carefully: lift budget in steps so Meta does not reset its learning, widen the radius or add a fresh lookalike, and feed new creative before fatigue sets in. Scale too fast and the cost per cover blows out overnight.
The money-wasting mistakes
Boosting instead of building. No pixel or Conversions API, so nothing is tracked. Targeting all of the city. Optimising for clicks. Running one ad until it dies. Pausing the second a campaign dips instead of reading why. We have seen every one of these drain a budget. We design the account to avoid them.
How it plays out for a real venue.
Illustrative numbers, but the structure is exactly how we run it.
Case study · Bar & kitchen, Fortitude Valley, Brisbane
The Loading Dock: from boosted posts to a tracked $9.40 cover
The Loading Dock is a 90-seat bar and kitchen off Brunswick Street. Wednesday to Friday ran hot. Tuesday was dead and Sunday lunch was worse. The owner had been boosting the odd post for fifty dollars a pop and could not tell you a single booking it had ever made. Spend over six months: about $1,800, results unknown.
We rebuilt the account in their own Business Manager, installed the pixel and Conversions API against their booking system, and tested a real booking until the event fired clean. Then we ran the three-campaign structure: an awareness campaign to a four-kilometre radius across the Valley, New Farm and Teneriffe, an offer campaign pushing a midweek share-plate-and-wine deal optimised for bookings, and a retargeting campaign to video viewers and site visitors.
Creative was their own phone footage, the kitchen pass on a Friday and the signature plate, cut into three angles and captioned for mute.
The retargeting campaign produced the cheapest covers by a distance, under five dollars each, because it spoke to people already warm from the awareness campaign. Tuesday now runs at two-thirds full most weeks. The owner can see exactly which campaign and which ad did it, because it is all in their account, which they own.
Everything in the managed build.
No vague "ad management". Here is the actual list.
- Business Manager, ad account and pixel set up in your name, owned by you.
- Meta pixel and Conversions API installed and tested against a real booking.
- Conversion events configured and ranked so bookings are what we optimise toward.
- The three-campaign structure built: awareness, offer-conversion and retargeting.
- Geo-radius, lookalike and custom audiences built from your own customer data.
- Creative direction, scripting and editing of your footage into tested ad variations.
- Weekly optimisation: budget shifts, audience tuning, frequency caps, creative refresh.
- Monthly reporting in cost per booked cover and return, in plain dollars.
- Compliance handled for any alcohol or age-restricted creative.
What happens, and when.
First 30 days
Account and tracking set up in your name, pixel and Conversions API tested on a live booking, first creative cut from your footage, and the three campaigns launched to a tight radius on a controlled test budget. We are reading early data, not scaling yet.
By day 60
We know what a cover costs. Winning audiences and ads get more budget, losers are cut, lookalikes are refined off the bookings coming in, and retargeting is producing the cheapest covers. The first clear cost-per-cover number is on the table.
By day 90
The campaigns are scaled to the level your covers justify, creative is on a refresh cycle so nothing fatigues, and you have three months of reporting in dollars. From here it is steady optimisation and pushing the spend that pays.
Run it yourself, or hand it over.
The Playbook (DIY)
Paid ads are one part of the full Packed Out system, written down with audience setups, ad templates and a tracking checklist. Module 10 walks you through running Meta ads yourself. A one-off $499.
Done for you
Our team builds, runs and optimises your Meta ads ongoing, alongside the rest of the system if you want it. You set the budget, we make it work.
Ads are one channel. Here are the others.
Owners ask us this about ads.
How much should a restaurant spend on Facebook ads?
For most independent venues, a starting ad budget of a few hundred to around a thousand dollars a month is enough to fill specific quiet services or push an offer, because we target a tight radius rather than a whole city. The right number depends on your margin, the seats you are trying to fill and the value of a cover. We set a budget where the return is clear, then scale only what is paying for itself.
What's the difference between boosting a post and running real Facebook ads?
Boosting is the blue button that shows your post to a broad, loosely targeted audience to chase likes. Real ads are built in Ads Manager with a specific objective, precise location and audience targeting, proper creative testing and a tracked outcome like a booking. Boosting can win you reach. Structured ads win you covers. We run the second kind and judge them on cost per result.
Can Facebook and Instagram ads actually drive bookings, not just likes?
Yes, when they point at a real action and are tracked. We send people to a booking page, an offer claim or a click-to-call, install the Meta pixel and conversions API so we can see what happened, and optimise toward bookings rather than cheap clicks. If a campaign is only buying likes, we change it. The whole point is covers in the book.
Do I need a big follower count or fancy creative for ads to work?
No. Ads run to people whether or not they follow you, so a small page is fine. And the creative that works for venues is usually simple: a genuine phone video of the food and the room beats a polished agency production most of the time. We test a few angles, keep what performs and refresh it before it tires out.
Is ad spend included in your fee?
No. Your ad budget goes straight to Meta from your own account, so you control it and own the data and the pixel. Our fee covers the strategy, build, creative direction, testing, optimisation and reporting. That keeps things clean: you can always see exactly what went to Meta versus what went to managing the campaigns.
Why do I need the Conversions API as well as the pixel?
Because the pixel on its own now misses a big share of what happens. Apple's privacy changes, ad blockers and cookie limits break browser tracking, so bookings get lost. The Conversions API sends the same events to Meta server to server, from your booking system, so a booking still counts even when the browser signal drops. Running both, with the data deduplicated, is the difference between Meta seeing most of your results and seeing half. We set this up and test it on a real booking before we spend.
How long before Facebook ads start producing bookings?
You will usually see bookings in the first couple of weeks, but the number that matters, a stable cost per booked cover, takes about 30 to 60 days. Meta needs time to learn who converts, and we need data before we scale. The first month is a controlled test on a tight budget. By day 60 we know what a cover costs, and from there we push the spend that is paying for itself. Anyone promising a flood of bookings in week one is selling you something.
Will you run ads for a venue that serves alcohol?
Yes. Bars, pubs and any venue with a drinks list are squarely what we do. Alcohol ads follow Meta's restrictions and Australian responsible-service rules, which means age-gated targeting and creative that stays inside the lines. We handle that compliance as part of the build, so you get the bookings without the account getting flagged. If you run a bar or pub, our bar and pub marketing page covers the wider picture, and our Done-for-you service runs the ads alongside the rest of the system.
The guides behind this service.
Want to understand the how before you hand it over? These free guides break it down, with real Australian numbers.
Facebook & Instagram ads for restaurants
The no-waste Meta ads setup: the only campaigns worth running, local targeting, and the cost-per-cover maths.
Read the guide →
The real cost of UberEats & DoorDash
Exactly what the apps take from every order, and the margin maths on going direct.
Read the guide →
Let's turn ad spend into covers.
A 15-minute call, no pitch. We'll look at what you're spending now and tell you straight whether Meta ads can fill your quiet services, and at what cost.