Instagram Insights gives you a screen full of numbers with almost no explanation of what any of them count. Most of the confusion comes from mixing two different kinds of figure: counts, which grow simply because your account grew, and rates, which do not. This page is the shared background behind the guides above. The full version is the Instagram analytics guide.
What Instagram actually measures, and where each figure lives
First, a prerequisite that catches people out: Insights only exists on a professional account. If your profile is set to Personal you will see likes and comments and nothing else. Switching to Creator or Business turns Insights on, and it does not backfill — the historical data starts accumulating from the switch, not from the day you joined.
Once you have it, the numbers sit in three separate places, and knowing which place you are looking at matters more than the number itself.
| Figure | What it counts | Where it lives |
|---|---|---|
| Accounts reached | Unique accounts that saw the content at least once. One person who sees it five times counts once. | Per post, and aggregated at profile level over a date window |
| Views (historically impressions or plays) | Total times the content was displayed or played, repeats included. Always the larger number. | Per post, and aggregated at profile level |
| Interactions | Likes, comments, shares and saves added together | Per post; broken out individually in the post’s own insights panel |
| Accounts engaged | Unique accounts that did any of those things | Profile level, over your chosen date window |
| Follows and unfollows | Gross additions and gross departures, not the net figure on your profile | Profile level, audience section |
Instagram has renamed several of these more than once, most visibly by consolidating plays and impressions under a single “views” label. Trust the label in your own app over any article’s — including this one — and check the definition Instagram shows when you tap the little information icon beside a metric. The guide to reach, impressions and views works through what each one is and is not good for.
A count is not a rate, and only one of them survives growth
This is the distinction the whole category rests on. A count — reach, views, likes, saves — scales with the size of your audience. If your followers double and everything else stays exactly as effective, every count roughly doubles. That feels like improvement and measures nothing.
A rate divides a count by the audience it came from, which strips the size out. Engagement rate is interactions divided by followers (or by accounts reached, which answers a subtly different question). Reach rate is accounts reached divided by followers. Because the denominator moves with you, a rate stays comparable to itself across a year in which your account tripled — which is exactly the property you need when you are trying to work out whether something changed.
The practical consequence: when reach falls but followers rose faster, your rate may not have moved at all. And when a brand asks what your engagement is, they are asking for a rate, because a rate is the only version of the number that is comparable across creators. What counts as a good engagement rate covers the bands, and the engagement rate calculator computes yours both ways and prints the arithmetic it used.
The most misunderstood point in this category
Views is the biggest number on the screen and the least informative one. It counts repeats, so a small, loyal audience rewatching a Reel inflates it exactly like a large audience seeing it once. Reach is the honest denominator for “how far did this go”, and interactions divided by reach is the honest answer to “did the people who saw it care”. A jump in views with flat reach and flat interactions usually means the same people watched twice, which is interesting but is not distribution.
The three numbers worth tracking monthly
- Engagement rate by followers. One figure per month, averaged over the month’s posts. This is the number a brand will ask for, and the one you should be able to quote without looking it up.
- Reach rate. Accounts reached divided by followers. Published creator reporting tends to put ordinary somewhere in the twenties to forties as a percentage, but treat that as a band rather than a threshold — reach rate is heavily format-dependent, and a Reels-heavy account that regularly reaches non-followers can exceed 100% without anything unusual happening.
- Net follower change. Not gross follows. Gross follows with the unfollows hidden is the single easiest number to feel good about while standing still. Feed the net figure into the follower growth calculator if you want to see where the current rate actually lands.
Three numbers, once a month, written down somewhere you will still have them in a year. That record is worth more than any dashboard, because it is the only benchmark that is genuinely about your account.
Four ways people misread their own Insights
- Celebrating a count. Impressions or views up 40% is not a result until you know whether reach and followers moved with it.
- Comparing against a public benchmark instead of your own last month. Industry averages are built from samples with unknown composition. Your previous month is a sample of exactly one account: yours.
- Reading a single post as a trend. Post-level variance on Instagram is large. One post that underperforms is noise; a month of them is a signal.
- Blaming timing first. Posting time is a real variable but a small one, and every published “best time” chart is an average over an audience that is not yours. Why every best-time chart is wrong for you explains the problem, and the best time to post calculator at least converts the question into your own audience’s time zones rather than someone else’s.