How to Measure Restaurant Social Media (and Tell If It's Bringing In Customers)
To measure restaurant social media, track three things separately: who saw your post, who took a step toward you, and who bought. Then test one special you already run against its own past numbers. You don't need new software or a new discount.
A video can do great online while your Tuesday stays slow. Looking only at views won't tell you why, and neither will looking only at sales. You need both, side by side.
Track three levels, not one number
We use these same three levels in Surge's monthly client reports:
| Level | The question | What to track |
|---|---|---|
| Attention | Did people see it and care? | Views, reach, watch time, shares, saves, follower growth, profile visits |
| Intent | Did anyone take a step toward us? | Menu clicks, order-link visits, calls, direction requests, reservation or booking inquiries, offer redemptions |
| Sales | Did it show up in the business? | Order counts, sales of the featured item, total transactions, total sales |
Attention still matters, because nobody orders from a restaurant they've never heard of. Just use the numbers that say the most. In January 2025, Instagram head Adam Mosseri said: "The top three signals that matter most for ranking are watch time, likes and sends. So when looking at your insights, pay close attention to average watch time, likes per reach, and sends per reach."
For a restaurant, watch sends closely. Mosseri also said sends are "slightly more important for unconnected content," meaning people who don't already follow you. That's the new customer you're trying to reach.
Test a special you already run
This is the simplest way to connect social media to sales. Say you run Monday wings:
- Pick one special. Use something already on your menu. You don't give up any margin on a new discount, and your old numbers are still comparable.
- Write down your baseline. From your POS, pull wing orders and total transactions for several comparable Mondays before you start.
- Promote it. Post entertaining videos that feature the wings, plus one direct post that says what the deal is and when to come in.
- Pull the same numbers for the Mondays during the campaign.
- Compare, and write down what else was going on. Weather, holidays and competing promos can all move a Monday. So can a big local event.
If wing orders rise and the rest of the menu holds steady, that's useful evidence. It's still evidence, not proof. Be suspicious of anyone who says otherwise, including us.
Give your posts something to measure
A funny video with no specific ask can build your audience, but it gives you very little to count. That's why at least 20% of the posts in our client calendars ask for one specific action: come in for this special, order this item, book this event. The rest are there to entertain people and earn their attention, and about half our videos do both by showing the food or the place inside something fun.
If none of your posts ask for anything, your intent and sales numbers have nothing to respond to.
Promo codes help, but know what they prove
The common advice is to give social media its own promo code and count redemptions. It's a good tool when you have a distinct offer. Many systems can handle it. Toast, for example, lets customers redeem codes on the POS, at a kiosk or through online ordering. Check what your own system reports before you start, so you aren't counting by hand.
Just know what the count means. A code proves a tracked purchase happened. It doesn't prove social media caused it, because a regular who was already coming in can use the code too. And a code means offering a discount, which the baseline test above doesn't require.
One free addition: have your staff ask, "How did you hear about us?" It won't give you a perfect number, but it can give you real customers' own words.
How to read your results
| What you see | What it likely means | What to try |
|---|---|---|
| Attention up, intent flat | People like the content, but nothing tells them what to do next | Add posts with one clear ask |
| Intent up, sales flat | People are looking, then stopping somewhere | Check your menu link, ordering page, hours and phone number |
| Attention, intent and sales all up | Your campaign may be working | Run it again on another special and see if it repeats |
| Attention flat | The content isn't earning a watch | Work on the first few seconds and the idea itself before you post more |
If you can't get sales data, that's okay. Report attention and intent, and don't make revenue claims. An honest partial picture beats a confident guess.
Want help setting this up?
We make entertaining social content for local businesses, and we'd rather show you what we can back up than promise a revenue number. If you want to set up a test like this for your restaurant, book a call with Surge.