There is no shadowban checker that works. Here is how often a healthy account posts a Reel that flops anyway
No tool can detect a shadowban. One Reel in eight normally lands under a third of its account median. Measured across 600 Reels, 24 September 2026.
What the numbers say
Start with the part nobody selling a shadowban checker will tell you: Instagram does not publish a shadowban status, there is no field for it in any API, and so nothing can read it. Every tool claiming to check one is inferring it from your reach, which you can already see. The useful question is different. How far does a Reel normally fall before the fall means something? Across 64 accounts with at least eight analysed Reels each, 600 Reels in total, 70 of them landed under a third of their own account's median views. That is roughly one Reel in eight, on accounts that were performing fine. Another 89 landed between a third and two thirds. So about one Reel in four normally comes in well under par, and the account is not being suppressed. It is just how the distribution looks.
One bad Reel is not evidence of anything. Two in a row is not either. What would be evidence is your whole distribution shifting down, which you can only see by comparing a run of recent Reels against your own older median, not against somebody else's benchmark.
Sample of 600, measured on 2026-09-24. Scrptly production database, queried 2026-09-24. 600 Reels across 64 accounts, restricted to accounts holding at least 8 analysed Reels so that each account's own median means something. Each Reel compared against the median of its own account.
The measurement
Each Reel measured against the median of its own account, so a large account and a small one can sit in the same table. Only accounts with at least 8 analysed Reels are included. The five bands are exclusive and hold all 600 Reels: 70 plus 89 plus 256 plus 85 plus 100. Find where your recent run sits. If most of it is still in the middle band, nothing unusual is happening.
| How the reel did against its own account | Reels measured | Roughly |
|---|---|---|
| Under a third of the account median | 70 | About 1 Reel in 8 |
| A third to two thirds | 89 | About 1 Reel in 7 |
| Around the median, two thirds to one and a half times | 256 | About 4 Reels in 9 |
| One and a half to three times | 85 | About 1 Reel in 7 |
| Over three times the median | 100 | About 1 Reel in 6 |
One Reel, beat by beat
Worth seeing what an ordinary Reel from a large account looks like, because the numbers are humbling and nobody calls it a shadowban. This is one real Reel from the corpus: 1,001,047 followers, seen 584,340 times. That is 0.58 times the following, so it reached well under half the account's own audience. It runs 86 seconds. It is a mid-series episode and it is structurally competent throughout. Reach under your follower count is completely normal and is not a penalty.
| At | What happens |
|---|---|
| 0:00 to 0:05 | Series intro: name, day number, quick scene setting for anyone arriving cold |
| 0:05 to 0:20 | A small situational update, then the confession that the series stopped six weeks ago |
| 0:20 to 0:50 | Why it stopped: the emotional toll, the specific triggers, the exhaustion. The longest beat |
| 0:50 to 1:10 | What happened instead: a break, some peace made with it |
| 1:10 to 1:20 | Where things stand now, stated plainly rather than resolved |
| 1:20 to 1:26 | Close by asking anyone who got this far to comment, then a sign off |
What to do next
- Do not buy a shadowban checker. There is no status for it to read, so whatever number it shows you has been inferred from reach figures you can already see for free in Insights.
- Work out your own median views across your last twenty Reels, then compare your recent run against that. A single Reel under a third of your median is ordinary, it happens to about one Reel in eight in this corpus.
- What is worth investigating is a sustained shift: most of your last ten Reels sitting in the bottom two bands when they used to sit in the middle. That is a pattern rather than a bad day.
- Check the boring causes before the conspiracy. A changed posting format, a switch of topic, music you do not have rights to, or a run of weaker openings will all move reach, and all of them are fixable.
- Two honest limits on the numbers above. This is 64 accounts, which is a modest sample, and they are accounts somebody chose to analyse, so they skew towards ones that were working. It is a fair guide to normal variance and not a clinical threshold.
- If you want your recent run scored rather than eyeballed, run your handle through Scrptly. It returns the hook score and structure for each Reel, which is where a sustained drop usually turns out to have started. The first analysis does not need an account.