Restaurant Social Media: What to Post and How to Run It

Q: How often should a restaurant post on social media?

There is no published, universal figure. Set cadence backwards from capacity: work out how much you can genuinely capture during service, select and approve in a normal week, then set the cadence slightly below that so a busy week does not break the routine. A venue posting twice a week for a year has a working system; a venue posting daily for a month and then going quiet does not. Consistency you can hold matters more than frequency you cannot.

Q: Why is my restaurant's Instagram not bringing customers in?

The fault is rarely the photographs. It is usually the step after the post: somebody becomes interested, taps through, and finds a profile or website that does not answer their next question. Check the path yourself on a phone - post, profile, link, booking - and see where you hesitate. Other common causes are posting that is not local enough to reach people who could realistically visit, no clear reason to come this week rather than at some point, and no routine for answering the questions that arrive in comments and messages.

The short answer to “what should we post” is: the things that are already happening in your building. The dish that comes back to the pass looking the way it should. The room at seven o'clock. The person who has worked your bar for six years. The reason to come this month rather than next.

The harder question is what happens after the post — and that is usually where restaurant social media quietly stops working. Most accounts are not failing because the photographs are bad. They are failing because the post is treated as the finished job rather than the first step: there is nowhere obvious for an interested person to go next, and no routine that turns a good week of service into a good week of content.

This guide sets out a content system an independent venue can actually run: what to capture, how to organise it around service, where each post should lead, and how to tell whether any of it is doing anything. Every factual claim about how a platform works is sourced to that platform's own documentation and linked, so you can check it rather than take my word for it. Where something is my judgement rather than a documented fact, I say so.

1. Why restaurant social media stops short

The short answer: a post is not the end of the job. It is one link in a chain that runs from something real happening in your venue through to somebody sitting at a table — and the chain usually breaks in the same place, at the step after the post, where an interested person cannot easily work out what to do next.

It is worth being precise about the chain, because “our social isn't working” describes at least four different faults. Somebody sees a plate of food on a Tuesday evening. They are mildly interested. They tap the profile. What they find there decides whether that interest turns into anything: a menu they can read, hours they can trust, a booking link that works on a phone, an address that makes sense. If any of that is missing, the interest evaporates — and no amount of better photography further up the chain will recover it.

The last stage is the one most venues skip entirely. Without a short, regular look at what happened, you are not running a content system; you are producing posts and hoping.

Diagram 1 — From a venue moment to a customer action

Where the chain usually breaks is stage five, not stage three.

A useful venue moment

Something genuinely happening: a dish going out, a delivery landing, a section of the room at its best.
Planned content

The moment captured against a pillar you already film every week, not invented on the day.
The post

Published with a caption that says what it is, where you are, and what someone can do about it.
Local discovery

Somebody nearby encounters it — through the feed, through search, or through someone who already knows you.
A destination that answers

Profile, Google Business Profile or website: readable menu, current hours, a working booking or order link.
Booking, visit or direct order

The action you actually wanted. It happens here, not on the post.
Weekly review

Ten minutes looking at what was captured, what was published, what was asked, and what changed.
↑ and the review decides what gets planned next week

Read it this way: if you are unhappy with the result, work backwards from stage six rather than forwards from stage one. Most venues put their effort into stages one to three and have never checked stage five on a phone.

2. What to post: four pillars you can sustain

The short answer: stop looking for ideas and start looking for categories. A pillar is something you can film again next week without inventing anything, because it is already part of the operation.

Ideas run out. Pillars do not. The difference matters on a wet Wednesday when nobody has planned anything and service starts in an hour: with pillars, there is always something to capture, because the pillars describe things that happen anyway.

Diagram 2 — Four content pillars for an independent venue

Each one is a thing your venue already produces. Rotate through them rather than starting from a blank page.

Food & drink

A dish plated properly, shot at the pass in real light
A special, with the reason it exists this week
Where an ingredient came from, and who from
How something is made, when the making is interesting

The room

The space at the hour it looks best
A corner people ask to sit in, and why
Changes: new tables, a refit, the terrace opening
What the room is like when it is quiet, not only when it is full

People & service

Who is actually cooking or pouring, named if they are happy to be
A skill being done well — carving, pulling a pint, folding pasta
A new starter, a long server, a supplier who turns up every week
How you do something differently from the place next door

Reasons to visit now

What is on this week that will not be on next week
Opening changes: bank holidays, late kitchens, last orders
Set menus, tastings, quiz nights, supper clubs
Practical answers: parking, step-free access, dogs, large tables

The fourth pillar is the one most venues under-use. It is also the one closest to a decision: “reasons to visit now” is the difference between somebody admiring your food and somebody booking a table for Friday.

Four is not a magic number — it is the largest set most small teams can keep going. If you can only sustain three, run three. A pillar you abandon after a month was never a pillar.

3. How often, and on which platforms

The short answer: there is no universal posting frequency, and anyone who gives you one has not seen your rota. Set cadence from what you can genuinely capture and approve each week, then hold it.

I am not going to print a number here, because no platform publishes one and the honest answer depends on your kitchen, not on best practice. What I would say is this: a venue that posts twice a week for a year has a working system, and a venue that posts daily for three weeks and then stops for two months does not. Consistency is a function of capture capacity. Decide what you can film during service without anybody stepping away from a section, and set the cadence just below that.

On platforms, the useful question is not “which platform is best” but “where are the people who could realistically walk through my door?” A neighbourhood pub with a settled local trade and a city-centre restaurant taking bookings three weeks out are not solving the same problem. Pick the smallest number of platforms you can post to properly. Two done well will beat five done thinly, and running an account you have abandoned is worse than not having one — an out-of-date profile is a live signal that nobody is home.

It is also worth retiring the idea that there is a single algorithm to beat. Instagram's own explanation is explicit: Instagram doesn't have a singular algorithm that oversees what people do and don't see on the app, and each part of the app — Feed, Stories, Explore, Reels, Search — uses its own algorithm tailored to how people use it.[2] That post is dated 31 May 2023, and platforms change, so treat it as a description of an approach rather than a permanent specification. The operational point survives either way: “the algorithm” is not one thing you can outwit, so build for the person, not the ranking system.

4. Turning service into content without a second job

The short answer: capture during service, decide afterwards. The reason social media feels like an extra job is that most venues try to create content at a time when nothing is happening, instead of collecting it when everything is.

The workflow below is deliberately small. It assumes one person holds the account, one person approves, and nobody has an afternoon free. The two steps that get skipped are approval and review — approval because it feels like bureaucracy until something goes out that should not have, and review because nothing forces it.

Diagram 3 — A weekly content workflow for a small team

Seven steps, one loop. The owner labels are a suggestion — what matters is that each step has exactly one name against it.

STEP 1
Plan

Fifteen minutes. Which pillars this week, and what is on that gives someone a reason to come.
Owner or GM
STEP 2
Capture in service

Phone, real light, no staging. A few seconds at the pass, not a shoot on a closed day.
Whoever is on
STEP 3
Select

Most of what you captured is not usable. Choose the few that are and discard the rest.
Account holder
STEP 4
Approve

One person, one channel, one deadline. Checks the facts: prices, allergens, times, names, permission to appear.
Owner or GM
STEP 5
Schedule and publish

Queued in one sitting so publishing never competes with a busy service.
Account holder
STEP 6
Respond

Community management is the job. Questions about booking, dietary requirements and opening hours are enquiries, not comments.
Named on the rota
STEP 7
Review

Ten minutes. What was captured, what went out, what was asked, what needs fixing on the destination.
Owner or GM

Then it starts again. The review is what feeds the next plan, which is why it is a loop rather than a list. Steps two and six are the only ones that touch a live service; everything else can sit in a quiet hour.

One practical note on approval: it is not there to slow anybody down. It is there because a menu price, an allergen, a closing time or a member of staff who did not want to be filmed are all things that are cheap to check before publication and expensive to correct afterwards.

5. Making your profile a destination, not a dead end

The short answer: assume every post ends with somebody tapping your profile, and build the profile for that person. Their questions are boring and predictable: what is the food, when are you open, where are you, and how do I book?

Two things are worth setting up properly, because both are documented and both are frequently left undone.

An action button on your Instagram business profile. Instagram's help documentation lists the buttons a business account can add — Book now, Reserve and Order food — and states plainly that Only one action button can be displayed on your profile at a time.[1] It also notes that you need an existing account with a partner to select them, and that the number of providers is limited and availability varies by region.[1] So this is a decision, not a checkbox: if you can only show one, show the one that matches how you actually take trade. A restaurant that lives on bookings and a takeaway that lives on orders should not be choosing the same button.

Social links on your Google Business Profile. Google's documentation says you can add one link per platform — Facebook, Instagram, LinkedIn, Pinterest, TikTok, X and YouTube — from the Contact section of your profile, and that You can add one link per social media platform.[3] Google also notes the feature is currently available in select regions, so check your own profile rather than assuming.[3] This is the join between the two halves of local discovery: the person who found you on Google can see the account, and the person who found the account can reach everything else. If you want the fuller picture of that side, I have written it up separately in the guide to how restaurants get found on Google, Maps and AI search.

After that it is mostly housekeeping, and it is housekeeping that decides outcomes. Send the link in your bio to the thing you want people to do — not to a homepage that makes them hunt for it. Make sure the menu on the other end is readable on a phone rather than a PDF that has to be pinched and dragged. Keep hours accurate, especially around bank holidays. And check the whole path yourself, on your own phone, as a stranger would: tap the profile, tap the link, try to book. Most venues have never done this, and it takes ninety seconds.

If the destination itself is the weak link — a site that is slow, hard to read on a phone or missing a working booking path — that is a website problem rather than a content problem, and it is worth fixing first. What that involves is set out on the services page, and the local-search side is covered under local SEO for restaurants.

6. What to measure, and what not to

The short answer: separate the things that prove you are running the system from the things that suggest somebody moved along the journey. Both are useful. Only one of them tells you anything commercial, and follower count is in neither column.

The trap is that the easiest numbers to find are the least informative. Activity signals tell you whether the routine is alive — genuinely worth knowing, because most content systems die quietly rather than dramatically. Journey signals tell you whether anybody moved towards the venue. Review the first weekly and the second monthly, because activity changes week to week and journeys do not.

Diagram 4 — Activity signals and journey signals

Two different questions. “Are we doing the work?” and “did it lead anywhere?”

Activity signals

Review weekly · is the system alive?

Posts published against the plan
Assets captured during service
Questions and messages answered, and how quickly
Profile kept current: hours, menu link, action button

Journey signals

Review monthly · did it lead anywhere?

Website visits arriving from social
Menu and booking-page visits
Contact and directions intent, where your tools expose it
Trackable direct actions — a booking link, a code, a form

Not everything is measurable, and pretending otherwise is how people end up trusting the wrong number. Google's own documentation states that Google does not currently provide performance metrics or click tracking for social media links on a Business Profile.[3] Some of the most valuable movement in hospitality — somebody who saw a post in March and walked in with three friends in May — will never appear in any dashboard. That is an argument for asking people how they heard about you, not for inventing a figure.

One measure I would treat with suspicion is follower count on its own. It is the number every platform puts in front of you and the one least connected to a table. A venue with a few thousand genuinely local followers is in a stronger position than one with ten times that spread across the country, and neither number tells you what happened on Friday.

7. What social media cannot fix

The short answer: content is a distribution problem. If the underlying offer, delivery or path to booking is broken, better distribution makes the problem more visible, not smaller.

This is the part most guides leave out, and it is the part worth reading twice. Social media cannot fix:

Unclear positioning. If you cannot say in one sentence what your venue is for and who it is for, no content plan will resolve it — every post will pull in a slightly different direction. Where that shows up most is in how the place looks and reads, which is brand work rather than content work.
Operational delivery. If the food arrives late, cold or inconsistent, more people seeing it means more people finding that out. Content accelerates whatever is already true.
A broken booking or order path. A link that fails on a phone, a booking widget that will not take a table for two on a Tuesday, or a menu locked in a file. This is the most common and the most fixable.
A website that does not answer the question. Interest arrives, finds nothing useful, and leaves. The site is where the decision gets made.
An empty room on a Tuesday in February. Some demand problems are structural — trading pattern, location, pricing, the offer itself — and are solved on the floor and in the P&L, not on a phone.

There is also a limit worth naming on the other side: social media is rented ground. The account is not yours, the reach is not guaranteed, and the rules change without notice. That is a reason to keep building the things you do own alongside it — your website, your Google Business Profile, and a list of people who have already been in, which is what email marketing for restaurants is for.

8. Running the venue and filling the venue

The short answer: the content that works is a description of a venue that is genuinely worth describing. The two halves are not separate projects, and the join between them is what a customer experiences as a journey.

Diagram 5 — The two halves have to meet

Neither half works alone. A well-run venue nobody can find, and a well-marketed venue that disappoints, fail in different ways.

Running the venue

Real food, service, people, standards, the operation behind them.

Filling the venue

Content, local discovery, the website, a booking path that works.

A clearer customer journey

Someone can find you, understand you, and act — and what they find is what they get.

This is not a promise about results. It is a description of where the work has to sit. Content built on top of an operation you are not proud of will be found out faster than content built on one you are.

A practical starting checklist

If you want somewhere concrete to begin, in order:

Walk your own path on your own phone. Find a recent post, tap through to the profile, tap the link, and try to book or order. Note every point where you hesitate.
Fix the destination before the content. Readable menu, accurate hours, working booking or order link. This is almost always the cheapest improvement available.
Set your action button to match how you actually take trade, remembering only one can show at a time.[1]
Add your social links to your Google Business Profile, one per platform, if the feature is available on your profile.[3]
Choose three or four pillars you can genuinely film every week, and write them down where the team can see them.
Name the roles. Who captures, who selects, who approves, who replies. One name against each.
Set a cadence you can hold in a bad week, not a good one.
Put a fifteen-minute plan and a ten-minute review in the diary as recurring events, on the same day each week.
Start asking people how they heard about you, and write the answers down. It is the least sophisticated measurement available and often the most useful.

If you would rather have this run properly rather than run it yourself, that is what the social media management service covers — planning, capture, publication and the reporting around it. If the underlying picture is the issue — the site, the menu, how you are found — the services page sets out what that involves, and you can get in touch here. Everything else I have written is on the blog.

Frequently asked questions

How often should a restaurant post on social media?

There is no published, universal figure, and I would be sceptical of anyone who gives you one as though there were. The practical way to set it is backwards from capacity: work out how much you can genuinely capture during service, select and approve in a normal week, then set the cadence slightly below that so a busy week does not break the routine. A venue posting twice a week for a year has a working system; a venue posting daily for a month and then going quiet does not. Consistency you can hold matters more than frequency you cannot.

Why is my restaurant's Instagram not bringing customers in?

In my experience the fault is rarely the photographs. It is usually the step after the post: somebody becomes interested, taps through, and finds a profile or website that does not answer their next question. Check the path yourself on a phone — post, profile, link, booking — and see where you hesitate. The other common causes are posting that is not local enough to reach people who could realistically visit, no clear reason to come this week rather than at some point, and no routine for answering the questions that arrive in comments and messages. Those are enquiries, and unanswered enquiries do not become bookings.

What should a restaurant post on Instagram?

Work from repeatable categories rather than one-off ideas. Four that most venues can sustain: food and drink as it actually leaves the pass; the room at the hour it looks best; the people and the skills behind the service; and concrete reasons to visit now — what is on this week, opening changes, set menus, practical answers about access, parking or large tables. The last category is the one closest to a decision and the one most venues under-use. If you can only keep three of the four going, keep three.

Which social media platforms should a restaurant actually use?

The smallest number you can post to properly, chosen by where the people who could realistically walk through your door already are. A neighbourhood pub and a city-centre restaurant taking bookings weeks out are not solving the same problem, so a single recommendation for both would be wrong. Two platforms done consistently will do more than five done thinly, and an abandoned account is worse than no account — an out-of-date profile tells a potential customer that nobody is paying attention. Whichever you choose, make sure it is linked from your Google Business Profile where that feature is available to you.

Can I add a booking link to my Instagram profile?

Yes, through an action button, with conditions. Instagram's help documentation lists “Book now”, “Reserve” and “Order food” as buttons a business account can add, states that only one action button can be displayed on your profile at a time, and notes that you need an existing account with one of its partner providers. It also says the number of providers is limited and availability varies by region, so check what is offered on your own account rather than assuming. Because only one can show, choose the button that matches how you actually take trade — reservations or orders — rather than adding whichever appears first.

What should I expect from a restaurant social media manager?

Ask what they do in a normal week, not what results they promise. A serious answer covers: how content gets planned against agreed pillars; who captures material and when, given that the best material comes from live service; how approval works before anything is published; what cadence they will hold and on which platforms; who answers questions and messages, and how quickly; and what they report on, in terms of the customer journey rather than follower counts. Be wary of anyone guaranteeing bookings, reach or engagement — those depend on platform behaviour nobody controls. Be equally wary of a proposal with no approval step in it.

Sources

Instagram Help Centre, “About action buttons for Instagram business profiles” — help.instagram.com/122793804938499. ↑
Instagram, “Instagram Ranking Explained”, Adam Mosseri, 31 May 2023 — about.instagram.com. ↑
Google Business Profile Help, “Manage your social media links” — support.google.com/business/answer/13580646. ↑

Last checked: 10 September 2026. Every quotation above was taken from the linked page on that date. Platform features and their regional availability change — if you are relying on one, open the source and confirm it still reads the same way. Everything not attributed to a source above is practical advice rather than documented fact. No ranking, reach, engagement or booking outcome is promised or implied.