How often should a restaurant post on Instagram?

Short answer: three to five feed posts a week, plus stories on the days you're open. If that sounds like a lot, post three times a week and don't miss. Three a week held for a year beats seven a week held for a month, and it isn't close.

That second sentence is the whole article, really. But it's worth explaining why, because the reason changes what you post, not just how often.

Frequency is the wrong first question

Most owners ask about frequency because they've read that the algorithm rewards it. That's roughly true and also not very useful, because it skips the part that actually decides your outcome: whether the cadence survives a busy month.

Here's the pattern we see constantly. A restaurant decides to get serious about Instagram. Week one, seven posts — the dining room, the new special, a staff feature, a reel. Week two, five. Week three, a walk-in cooler dies and there are two posts. Week four, nothing. Month two, the account is dormant and the owner concludes Instagram doesn't work for restaurants.

Nothing about that failure was algorithmic. The cadence was set at the ceiling of what was possible on a good week, and restaurants don't have good weeks reliably.

Pick the number you can hold on your worst week, not your best one. That number is usually three.

What consistency actually buys you

Two things, and they're different.

The first is distribution. Platforms decide what to show partly based on what recent signal they have about you. An account that posts steadily gives the system something to work with continuously. An account that posts in bursts gets evaluated cold each time it wakes up. You're not being punished for the gap — you're just being re-guessed at.

The second matters more and gets discussed less: familiarity. A person who has seen your restaurant in their feed nine times over three months walks past your door differently than someone seeing it for the first time. That effect compounds only with repetition over time. Nine posts in one week and nothing after does not produce it. Nine posts spread over three months does.

That second effect is the one that actually fills tables, and it's purely a function of duration. Which is why the cadence you can sustain beats the cadence that sounds impressive.

A cadence that survives a bad week

Something like this works for most independent restaurants:

Three feed posts a week. One food, one people, one anything-else. The "anything else" slot is what keeps the account from looking like a menu catalog.
Stories on operating days. Low effort by design — a prep shot, a delivery arriving, the specials board. These aren't productions. If a story takes more than thirty seconds to shoot, it's the wrong story.
One short video a week, if you can. If you can't, drop this before you drop the feed posts. Video is worth more per unit and costs more to make, so it's the first thing that should flex.

Notice the ordering. When the week goes sideways, you cut video first, stories second, feed posts last. Decide that hierarchy now, while it's calm, because you won't decide it well at 7pm on a Friday.

Batch it, or it won't happen

The single highest-leverage change most restaurants can make isn't posting more. It's separating the day you make content from the day you post it.

Shoot once. Fifteen minutes with a phone during a quiet stretch — the prep, the pass, a plate in good light near a window, one shot of a person who works there. That's most of two weeks of material. Then posting becomes a two-minute scheduling task instead of a daily creative decision made under pressure.

Content dies at the "what do I post today" step, every time. Batching deletes that step.

When more is actually right

There are real cases for posting daily or more:

You're opening, or reopening, and need to build an audience from nothing fast
You run genuinely daily specials and the account functions as an operational channel people check
You have someone whose actual job includes this

If none of those apply, more frequency mostly buys you burnout and a slightly larger archive of posts nobody saw.

How to know if it's working

Don't judge by likes. Judge by these, in order:

Reach among people who don't follow you. This tells you whether the work is finding anyone new. It's the only number that speaks to growth.
Saves and shares. Someone saving a post is signalling intent far more strongly than someone liking it.
Profile visits, then taps to your directions or booking link. This is the actual funnel. Everything upstream is a means to this.

Give it sixty days before you draw conclusions, and ninety before you draw firm ones. Consistency compounds, which is another way of saying it looks like nothing is happening right up until it doesn't.

The honest caveat

Posting three times a week to an account with nothing worth looking at won't do much. Cadence is a multiplier on quality, not a substitute for it. If your photos are dark phone shots of plates under warm restaurant lighting, fixing that will do more for you this month than doubling your posting rate.

Start there. Then hold three a week for a year.