Social Media Marketing
Instagram analytics: what the numbers mean and what counts as good in 2026
Instagram tells a business account quite a lot: how many people saw a post, how many saved it, where profile visits came from and whether anyone clicked the link. Yet many business owners open Insights, stare at the numbers for a moment and close the app none the wiser.
That's not on you. There are a lot of numbers, some of them changed both name and meaning in 2025, and the claims about what counts as a "good" rate contradict each other. Let's go through what the numbers actually tell you, what a realistic level looks like for a small business, and where Instagram's own analytics run out.
Views replaced impressions in 2025
First, a quick staleness check: if a guide tells you to track impressions, it's out of date.
On April 21, 2025 Meta removed impressions and video plays and replaced them with a single metric, views. The change covers everything: posts, reels, stories and your profile (Meta's changelog).
The difference between views and reach is worth memorizing:
- Views count every look. If the same person watches your reel three times, that's three views.
- Reach counts every person once, no matter how many times they come back.
The practical consequence: if you compare this autumn's numbers to early 2025 or older, you're not comparing the same metric. A high view count doesn't necessarily mean your audience grew. It can mean the same people are rewatching.
Where to find the numbers
Insights are only available on a professional account. In the app, go to your profile, then Professional dashboard, then Insights; each post also has its own numbers under View insights.
If your account is still personal, convert it first. It's free and removes nothing, and the walkthrough is in our guide to connecting your Instagram and Facebook accounts.
The six numbers worth knowing
Instagram shows dozens of figures. For a small business, these six cover everything that matters:
| Metric | What it tells you | What you use it for |
|---|---|---|
| Views | How many times your content was looked at, repeats included | Comparing how well content pulls |
| Reach | How many different people saw it | The real size of your audience |
| Interactions | Likes, comments, shares and saves | What resonates |
| Profile visits | How many were curious enough to check who you are | Depth of interest |
| Link taps | How many moved on to your site or shop | From social to business |
| Follower growth | The direction of your audience week over week | The long trend |
Two of these deserve a closer look.
Saves and shares are the heavyweight interactions. They tell the algorithm your content was worth keeping or passing on, and that same signal decides who Instagram shows your content to next. There's more on this in our guide to how the Instagram algorithm works.
Profile visits and link taps are where likes turn into business. A salon's post can collect a modest number of hearts and still book three appointments, if the right people clicked the link. That's why these two belong next to interactions in every review.
What counts as a good engagement rate
Engagement rate is a post's interactions divided by your follower count. It's the most common comparison number, which is exactly why so much shaky information circulates about it.
Rival IQ tracks this in a yearly benchmark report. The 2026 edition puts the median for business accounts at 0.30% per post, down from 0.36% a year earlier. Higher education and sports sit at the top, retail and beauty near the bottom (Rival IQ 2026).
Before you panic about your own number, two caveats:
- The benchmarks don't agree. Socialinsider gets a median of roughly 0.48% with the same formula, because its sample is different. Both numbers are correct within their own data.
- Small accounts usually score higher. An account with a hundred followers hitting 2% means two people liked a post, and that's typical, because a small audience is often an engaged inner circle.
Compare against yourself, not the average
The most useful benchmark is your own previous month. If your engagement climbed from 0.4% to 0.6%, you did something right, whatever some list's "good rate" claims.
A weekly routine that's enough
Using analytics and business results go together in the research, too: in a survey of 334 small and medium-sized businesses, the use of social media analytics went hand in hand with better firm performance (Journal of Asia Business Studies). You don't need a data team for this, just a regular slot in the calendar. Fifteen minutes a week covers it:
- Sort the week's posts by interactions. Which one collected the most saves and shares? That's what your audience wants more of.
- Check reach. Did one post reach clearly more people than the rest? Work out what was different about it: the topic, the format or the timing.
- Pick one thing to repeat next week and one thing to drop. Without this step, the first two were entertainment.
Analytics earn their keep only when they change what you do next week. A report nobody acts on is a ritual.
Where Instagram's own analytics run out
Insights have three limits that every growing account eventually hits:
- The 90-day window. Account-level numbers go back at most 90 days, and older data drops out of sight. By December you can no longer see what worked in August.
- One channel at a time. Facebook, TikTok and the rest each keep their numbers in their own app, so the full picture has to be assembled by hand.
- A phone screen. Comparing longer periods or reading trends on a small display is a struggle.
This is the gap publishing tools fill. Steadybeat shows per-post numbers from all your channels in one view, draws follower growth over a full year and uses AI to point out what worked in your content. The three questions of the weekly routine get answered from a single screen.
Summary
Instagram analytics fit into four sentences. Views replaced impressions in 2025, so don't compare against old numbers. Six metrics are enough: views, reach, interactions, profile visits, link taps and follower growth. The median engagement rate is 0.30%, so a number that looks modest is probably normal. And what matters most is a weekly routine that turns the numbers into next week's decisions.
All your channels' numbers in one view
Steadybeat gathers your post performance from every channel, tracks follower growth and uses AI to tell you what worked.
Steadybeat – publish once, be seen everywhere.
Sources
Figures checked in September 2026.
- Meta: Graph API v22 announcement and the Instagram platform changelog. Impressions and plays removed on April 21, 2025.
- Rival IQ (Quid): 2026 Social Media Industry Benchmark Report. Medians come from the report's own sample, and different providers' samples produce different medians, so treat the figures as orders of magnitude.
- Journal of Asia Business Studies: a survey of 334 Thai SMEs linking the use of social media analytics to firm performance.
Read also
How Often Should You Post on Social Media? The Usual Advice Is Misleading
The famous Buffer 5× engagement stat is real, but easy to read wrong. Correlation, causation, and what consistency actually does for you.
How much does social media marketing cost? A 2026 pricing guide for small businesses
Agency €800–2,500/mo, freelancer €1,300/mo, in-house hire €4,500/mo, or do it yourself with a tool? We break down what social media marketing actually costs a small business.
Social Media Tool Pricing 2026
Why are you overpaying? We compared the real costs of the most popular social media tools.