We measured 533 Instagram accounts: followers predict almost nothing

Straight answer: across 533 Instagram accounts we track (each with at least 1,000 followers and at least 5 posts in the last 180 days), the median account gets views equal to 55% of its follower count. That single number is almost useless on its own, because the spread around it is enormous — the bottom quartile gets under 18%, the top quartile gets over 169%, and 35% of accounts get more views than they have followers. In practice this means follower count tells you very little about reach, and any rate card, media kit or outreach list built on followers alone is pricing noise.

Method (so you can argue with it)

We took every Instagram account in our index with at least 1,000 followers and at least 5 posts published in the trailing 180 days — 533 accounts. For each one we computed the MEDIAN view count of those posts, not the average. The median matters: a single 20M-view outlier drags an average up by an order of magnitude and tells you nothing about what the account does on a normal Tuesday. Then we divided that median by the account's current follower count. We call that ratio reach-per-follower.

• Sample: 533 Instagram accounts, ≥1,000 followers, ≥5 posts in 180 days.
• Metric: median views of those posts ÷ current follower count.
• Why median, not mean: one viral post inflates a mean beyond usefulness. The mean ratio in our sample is 5.14 — nine times the median. That gap IS the problem with follower-based math.
• Limitation, stated plainly: this is our crawl, not a census of Instagram. It skews toward accounts someone has looked up. Treat the shape as directional and the named examples as literal.

The distribution

• 25th percentile: 0.18 — this account gets 18 views per 100 followers.
• Median: 0.55 — 55 views per 100 followers.
• 75th percentile: 1.69 — 169 views per 100 followers.
• Mean: 5.14 — dragged up by a long tail of breakout accounts, which is why nobody should quote the mean.

Read the first and third numbers together. Two accounts with 50,000 followers each: the one at the 25th percentile does about 9,000 views a post, the one at the 75th does about 84,500. Same follower count on the media kit. Nine times the actual reach. If you are paying both the same rate — and follower-based rate cards say you should — you are overpaying one of them by roughly 9×.

Named examples from the index

These are real accounts in our index, each with enough posts to make the median meaningful. Follower counts and medians are as measured on 6 August 2026.

Big following, small reach

• @kasemshoes — 190,780 followers, 3,138 median views across 41 posts. Reach-per-follower: 0.02.
• @bakerist_uae — 194,744 followers, 6,086 median views across 11 posts. Reach-per-follower: 0.03.
• @adollshome_3.0 — 166,851 followers, 7,242 median views across 24 posts. Reach-per-follower: 0.04.
• @ui.ux.jam — 173,957 followers, 10,402 median views across 22 posts. Reach-per-follower: 0.06.
• @kemal.mistik — 194,403 followers, 23,253 median views across 23 posts. Reach-per-follower: 0.12.

Small following, large reach

• @lincolnmolloy — 5,572 followers, 348,844 median views across 16 posts. Reach-per-follower: 62.6.
• @saam.ben — 14,548 followers, 847,749 median views across 19 posts. Reach-per-follower: 58.3.
• @whazianbran — 9,786 followers, 433,945 median views across 12 posts. Reach-per-follower: 44.3.
• @bhaaprebhaap — 46,027 followers, 1,945,089 median views across 15 posts. Reach-per-follower: 42.3.
• @sami.madeo — 5,595 followers, 207,836 median views across 13 posts. Reach-per-follower: 37.2.

A 5,572-follower account in our index reliably out-reaches a 190,780-follower account by 111×. Both would appear on the same influencer list, sorted by the wrong column.

Why the gap exists

Instagram stopped being a follower-graph distribution system and became a recommendation system. Reels are served to people who do not follow the account, chosen on watch-through and re-watch behaviour. A follower count is a record of past distribution: it tells you how many people once tapped Follow, not how many the algorithm will serve tomorrow. Accounts that bought followers, grew on a format they abandoned, or built an audience in a country whose feed they no longer match all keep the number and lose the reach.

The reverse is equally common. An account with 5,000 followers that consistently produces something the recommender likes gets served far outside its follow graph, every post. That account is a better placement than a dormant 200,000-follower account at any price — and it will usually quote less, because it is pricing off the same follower number everyone else is.

What to use instead

1. Pull the account's last 8–12 posts and take the MEDIAN view count. Ignore the top post; it is the one they will show you.
2. Divide by follower count to get reach-per-follower. Under 0.18 puts them in the bottom quartile of our sample. Over 1.69 puts them in the top.
3. Check consistency, not peaks: how far apart are their 25th and 75th percentile posts? A creator whose median is 40k and whose worst post is 30k is a safer buy than one whose median is 40k off a single 2M spike.
4. Price from median views, not followers. A view is the unit you are actually buying.
5. Re-check before every renewal. Reach-per-follower moves; the follower count almost never goes down.