Social Media Metrics for Small Business: The 3 Numbers That Matter

To tell if your social media is working, watch three numbers: how many people saw your posts, how many of them reacted, and how many became customers. The third is the one that pays the bills, and the first two tell you why it's going up or down. You can ignore nearly everything else in your insights screen.

This choice of three is our judgment, not a law or a study. Nobody has proved that these are the three right numbers for every business. We chose them because a busy owner can check them in minutes, and each one points to something you can do. Below we explain the reasoning so you can disagree with it.

What are KPIs for social media?

KPI stands for key performance indicator. It's a number you've decided to watch because it tells you whether something is working. A social media KPI might be how many people a post reached, or how many calls came in from your profile.

The word that matters is "key". Instagram, Facebook and TikTok will happily show you dozens of numbers. If you watch all of them you're really watching none, because you'll always find one that went up. A big company with a marketing team can look at many. A salon owner between appointments needs a few that lead to a decision.

So before the numbers, one question: what do you want social media to do for you? For most local businesses the answer is simple. You want more people nearby to know you exist, to remember you, and to book or walk in.

Each of our three numbers maps to a step in that:

1. Seen by: did people see it?
2. Interactions: did they care?
3. New customers: did they do something?

Number 1: How many people saw it

This is reach, the number of different people who saw your post. Some apps call it accounts reached. Meta defines Instagram reel reach as the number of unique users who saw it at least once.

Why it matters: nobody can become a customer from a post they never saw. It's the top of the funnel, and the fairest way to compare one post with another.

Don't confuse it with views. Views are the times a post was shown or played, and repeat views count. One person watching your reel four times is four views and one person reached. Your app may show Views first, and it's fine to use if that's what you have. Just pick one and stay with it, so you're comparing like with like.

Also know that these are approximate. Meta marks its reach figure as estimated. Small differences between two posts mean very little.

How to read it: compare a post with your own recent posts, not with a famous account. What you're looking for is whether your typical post is reaching more people this quarter than last.

If it's low or falling, try:

• Posting more steadily. Our guide to how often to post gives a realistic number.
• Trying a different format. A short video may travel further than a photo, or it may not for you. Test it and see.
• Checking the first second or first line. If that's weak, people scroll past before they've really seen it.

Number 2: How many people reacted

Interactions are the things people do with a post: likes, comments, shares and saves. Meta describes Instagram's total interactions on a reel as likes, saves, comments and shares, less any taken back.

Why it matters: this is your best sign that a post meant something. It's a cheap, early signal, available within a day, long before any customer turns up.

Not all interactions are equal

The total is a good start, but it treats every action the same. We'd weigh them differently, and again, that's our view:

• Saves mean "I want this later". That's strong for a tip or a how-to.
• Shares mean someone put their own name behind you. A friend's recommendation is hard to buy.
• Comments take effort, especially ones with real words in them. A question from a local person is worth a lot, so answer it.
• Likes take a thumb and a second. They're nice, and they're the weakest signal here.

Working out a rate

A post seen by 2,000 people will usually get more likes than a post seen by 200, so raw totals favor posts that were already seen more. To compare fairly, divide:

Interactions ÷ people reached × 100 = engagement rate

That's the formula Hootsuite gives for engagement rate by reach. Here's a made-up example. A post is seen by 400 people and gets 22 interactions. 22 ÷ 400 = 0.055, or 5.5%.

What's a good number?

Nobody can really tell you, and you should be wary if someone is sure. These are the two published figures we found most useful, both from companies that sell social media software:

• Buffer looked at over 52 million posts made through its own tool. Its 2025 medians, using reach, were about 5.5% on Instagram, 5.6% on Facebook, 6.1% to 6.2% on LinkedIn, and 4.5% on TikTok. Buffer says these describe its own users, not universal benchmarks, and that it counts engagement differently per platform.
• Hootsuite says a good rate generally falls between 1% and 5%, without stating what that's based on. Its own figures for a platform's best-performing format are lower than Buffer's medians, for example 4.2% for Instagram carousels and 1.5% for TikTok video.

Those two don't agree, and they can't be compared directly, since each measures different things in a different group of accounts. Take away only a loose idea: low single digits to around 6% isn't unusual. Your own trend is a far better guide. Which of your own posts got double your usual rate? Make more like it.

If it's low or falling, try:

• Writing about customers' questions and problems more than about your business.
• Showing faces, hands and real work. Our list of 50 social media post ideas has plenty to try.
• Asking one easy question at the end of a post, and replying when people answer.

Number 3: How many became customers

This is the one that matters, and the one no app can fully give you. It's also the one most reports leave out.

Why it matters: you can't pay rent with reach. A post can do very well and bring no one in. A plain post about a Tuesday opening may reach few people and fill three chairs.

The catch: you'll never count this perfectly. Someone sees your reel in March, thinks of you in May, and searches your name. No tool will connect those two moments. So aim for a useful rough count, not a precise one.

Here are the ways to get that count, simplest first. You don't need all of them.

Ask every new customer

"How did you find us?" is the most useful marketing question there is, and it's free. Add it to your booking form, or ask at the counter. Keep a tally on paper or in your phone with a few columns: Instagram, Facebook, Google, a friend, walking past, other. People's memories aren't perfect, but a few months of tallies will show you a pattern.

Count direct messages and calls

Messages that say "do you have space on Saturday?" are customers raising their hand. Count the ones that lead to a booking or a sale. Tally these in the same place.

Check your Google Business Profile

People who see you on social media often look you up on Google before they call. Your profile's performance page shows calls, website clicks, direction requests, messages and bookings, where they apply to your business. It can't tell you which of them started on social media, and Google notes that the figures include people who came through its ads. Still, if these climb during a month you were posting steadily, that's a useful sign.

Use an offer or word only for social

Mention something in a post that isn't anywhere else, like "tell us you saw this on Instagram for a free add-on". Anyone who says it came from that post. Only do this if it's a real offer you're happy to honor.

Tag your links

If you have a website with Google Analytics, you can add UTM tags to your links, which show which campaign sent each visitor. Google has a tool that builds the link for you. This is the most technical option here, so skip it if you don't already use Analytics.

What we'd leave out, and why

These are the numbers we'd stop checking, or check rarely. You may reasonably weigh them differently.

• Followers. It's the first number everyone looks at, but it doesn't say whether those people live nearby or still see your posts. A slow rise is fine, and a local account with few followers can still fill a diary.
• Likes on their own. They're already counted in interactions, and they're the weakest signal.
• Views on their own. Repeat plays inflate them. They're a stand-in for reach when that's all you have.
• Profile visits and link taps. These are useful for finding out why customers aren't coming, but they don't need to be on your main list.
• Competitors' numbers. You can only see the outside of their accounts. You don't know what they spent on ads or whether any of it brings them customers.

When to check, and for how long

Checking every hour won't help, and we'd say it can do harm, because it pushes you to change things before they've had a chance to work. A rhythm we'd suggest:

• One day after posting: a quick look, only if you enjoy it. Was anything asked in the comments? Answer it.
• One week later: there's no fixed rule, but a week is a fair point to compare posts.
• Once a month: the real check, and it takes about 10 minutes.

Here's the monthly check:

1. Write down your typical reach per post this month, and compare it with last month.
2. Write down your typical interactions per post, and do the same.
3. Add up your customer tally: how many said they found you on social media?
4. Note your best post and your weakest post. Ask what was different.
5. Choose one thing to change next month. Just one, or you won't know what made the difference.

Don't judge the whole effort on a single post or a single month. A slow week, a holiday or a change in what the app favors can move your numbers without you doing anything. Give any real change a few months before you call it. That's a judgment too, as no study we trust gives an exact time.

Reading the three together

The value of having three numbers is what they tell you in combination. This is where they turn into decisions:

• Low reach, good interactions: people like what you make, but too few see it. Post more steadily, try more video, ask for shares.
• High reach, low interactions: people see it and move on. Rethink the content before you try to reach more people.
• Good reach and interactions, few customers: people enjoy your posts but don't know what to do next. Say what you offer, where you are and how to book. Check that your profile link and phone number work. Then ask yourself whether the audience is local at all.
• Low on everything, customers still coming: don't panic. If your tally shows friends and Google bringing people in, social media may be just a supporting player, and that's a fair answer.
• Low on everything for months, after real effort: look at whether your customers are on this platform, or spend your time somewhere else, like your Google profile or an email list.

Here's a made-up example of what this looks like. Say a nail studio averages 300 people reached per post and a 6% interaction rate, and 4 new clients in a month say they found it on Instagram. If each new client is worth $150 in profit over time, that's $600 a month. Now the owner can ask a real question: is that worth the hours spent, or what's being paid?

Where to find the numbers

Each app has its own place, and the screens change often, so look for these words:

• Instagram: Insights, on a post or in the professional dashboard. You'll need a professional account.
• Facebook: Insights in Meta Business Suite, or on your Page.
• TikTok: Analytics, in your business or creator tools.
• Google Business Profile: the Performance page.

A notebook or a simple spreadsheet is plenty for writing them down. If you'd rather not gather them by hand, a scheduling tool can pull many into one place, such as Buffer or Later, and you'd still keep your customer tally yourself. Another option is the software we make, IndieAM. It plans a month of marketing, writes your posts for you to approve, and publishes approved posts to connected accounts. For those accounts, it shows "Seen by" and "Interactions" per post at 24 hours, 7 days and 30 days, which are our first two numbers. It won't count your third for you, it runs no ads, and it doesn't reply to comments, so it's not for you if you want someone handling your community or your ad reporting. Our prices are on the home page.

Start with a tally

If you do just one thing after reading this, start tracking number three. Put a sheet of paper by your till or a note on your phone, and ask every new customer how they found you. In three months, you'll know more about whether your social media is working than any insights screen can tell you.