Three jobs, not one
Restaurant marketing fails when it is treated as one activity called "social media". It is three distinct jobs:
- Discovery — being found by people deciding where to eat.
- Proof — convincing them the food and experience are worth the trip.
- Repeat — bringing back people who already liked you.
Most restaurants over-invest in the second and neglect the first and third entirely.
Job 1: Discovery
Discovery is mostly technical and operational, not creative.
- Business profile: correct category, cuisine tags, hours including holidays, menu, and price range. Wrong hours are a direct revenue loss and a review risk.
- Photographs: current, well-lit, showing food as served, the interior as it actually looks, and the exterior so people can find the door. Update seasonally.
- Menu online, in text rather than as an image, so it is readable by search engines and assistants.
- Delivery platform listings kept in sync with the real menu and pricing.
- Reviews flowing steadily, with responses.
A restaurant with excellent food and stale listings loses to a mediocre one with current photographs and correct hours.
Job 2: Proof
Proof is where content earns its keep. What performs:
- Short vertical video of food being prepared and served. Movement and sound outperform styled stills.
- Guest reactions captured genuinely.
- The people — chef, staff, owner. Faces build familiarity that a plated dish cannot.
- Behind the scenes — sourcing, prep, early mornings.
- User content, reshared with credit, which is the most credible proof available.
Write captions that include what people actually search: dish names, cuisine, locality, occasion. Social search is real, and captions are the index.
Job 3: Repeat
This is the neglected one, and the cheapest revenue in the business.
Build a guest database using: wifi login, reservation records, birthday club sign-ups, delivery orders, feedback forms, and loyalty enrolment. Capture a phone number with consent and a date.
Then use it deliberately: a message when a favourite returns to the menu, a birthday offer, a festival menu announcement, a quiet-Tuesday incentive to your most frequent guests.
A restaurant with three thousand contacts and a monthly message has a demand lever competitors do not.
The weekly operating rhythm
Systems beat inspiration. A workable rhythm:
- Monday: review last week — covers, sources, reviews, delivery ratings.
- Tuesday: capture content in one batch. One session, several assets.
- Wednesday: schedule the week's posts and stories.
- Thursday: respond to every review and message from the week.
- Friday: push the weekend offer or event to the guest database.
- Sunday: photograph anything new; check listing accuracy.
Two to three hours a week, done consistently, outperforms a heroic monthly effort.
Offers that build rather than erode
Discounting trains guests to wait for discounts. Better structures:
- Off-peak incentives that fill empty capacity rather than discounting full nights.
- Bundles that raise average spend rather than cutting price.
- Limited-time menu items that create urgency without margin damage.
- Occasion packages for birthdays, anniversaries, and small groups.
- Loyalty that rewards frequency rather than a blanket percentage off.
Reviews as an operating system
- Ask at the table, at the moment of visible satisfaction.
- Respond to every review, especially critical ones, briefly and without argument.
- Read them for operational signal: repeated mentions of wait times or a specific dish are free consulting.
- Never buy reviews.
Measuring it
- Covers by day and daypart.
- New versus repeat guests, if your system can distinguish them.
- Review volume and average rating over time.
- Database size and message engagement.
- Delivery platform ratings and order volume.
- Cost per new guest from any paid activity.
The failure mode to avoid
The classic pattern: a burst of posting for two weeks, silence for a month, then panic promotions when a quiet week arrives. Panic discounting damages margin and trains guests badly.
A system that runs at a modest constant level produces more covers and less stress than intensity followed by neglect, and it survives the weeks when the kitchen is too busy for anyone to think about marketing.
