Instagram Reels not getting views suddenly? Here is how to diagnose it

When a Reel that would normally do 40,000 views does 900, the instinct is to assume Instagram has throttled your account. That is almost never what happened. In the overwhelming majority of sudden drops, your followers are still seeing your posts exactly as before — what stopped is distribution to people who don't follow you. Instagram ranks connected and unconnected audiences with different systems, and only one of them is easy to lose.

That distinction is the whole diagnosis, and you can confirm it in about five minutes inside Insights. Work through the procedure below before you change your posting schedule, rewrite your hashtags or delete anything. We sell Instagram views, so you should read the section on buying them with that in mind — the short version is that it treats a symptom and cannot treat four of the six causes.

Run this diagnostic before you change anything

This takes five minutes and it tells you which of two completely different problems you have. You need a professional account — Creator or Business — for Insights to be available at all.

1. 01Open the Reel that underperformed, tap the view count, then tap View insights. You want the full breakdown, not the summary on the grid.
2. 02Find the Views panel and expand the split between Followers and Non-followers. This is the single most diagnostic number on the page and most people never open it.
3. 03Write down both figures. Then do the same for your previous ten Reels, so you have a baseline rather than a feeling.
4. 04Compare. If follower views held roughly steady and non-follower views collapsed, you have lost recommendation eligibility. Instagram is still showing the Reel to people who follow you and has stopped showing it to everyone else.
5. 05If both fell together, it is not an eligibility problem. It is the content — the Reel did not hold attention, and your own audience told Instagram that first.
6. 06Now check Settings → Account Status. This is where Instagram tells you directly if your account or a specific post is not eligible for recommendations, and it carries an appeal path. It has existed since December 2022 and it is the closest thing to an official answer you will get.
7. 07Check seconds watched and average watch time on the underperforming Reel against your baseline. If the average dropped below three seconds, the opening is the problem regardless of anything else.

Symptom, cause, check, fix

Match what you are actually seeing to the row that describes it. The right-hand column is the only action worth taking for that pattern.

The six causes that explain nearly all of it

These are ordered roughly by how often they are the answer in 2026, not by severity.

1. The original-content rule, effective 30 April 2026

This is the highest-stakes change Instagram has made to distribution in years, and it is the most common cause of a sudden collapse right now. Accounts that repeatedly repost other people's work are no longer eligible for recommendations across the app — and as of that date the rule covers photos and carousels, not just Reels. Your followers still see your posts. Nobody else does.

What counts as unoriginal is broader than most people assume. Straight reposts obviously qualify. So do edits where you have added a watermark, edits where you have only changed the speed, and screenshots of someone else's post — even when you credit them. Crediting the original creator is good manners. It is not a distribution exemption.

The fix is slow and there is no way to speed it up. Stop posting other people's material, publish original work consistently, and give it several weeks. Eligibility is assessed on a pattern of behavior, so a single original Reel does not reverse a pattern of forty reposts.

2. A visible watermark from another platform

Be precise about this one, because the internet gets it backwards. Third-party editing apps are not penalized. Editing in CapCut does not cost you reach, and Adam Mosseri has said so directly. What costs you recommendation eligibility is a visible watermark from another platform burned into the frame — the logo and username overlay that a downloader leaves behind.

The practical fix is to export a clean file rather than saving from another app's share sheet. If the watermark is already in the video, cropping it usually damages the frame more than it helps; re-export instead.

3. Engagement bait

This is the most under-known real penalty in the category, and it is not subtle. Instagram states the rule plainly.

"Comment a word for the link" is the specific pattern. So is "tag three friends", "share this to your Story", and "double tap if you agree". The irony is that these captions are recommended constantly by growth accounts as a way to increase reach, and they do the exact opposite: they remove the post from recommendations entirely, which is where most of your non-follower views come from.

Open-ended questions are fine. "What would you have done here?" is a question. "Comment YES if you agree" is an instruction to perform an action, and that is the line.

4. A Recommendations Guidelines violation

Content can be perfectly allowed on Instagram and still not be recommendable. The Recommendations Guidelines are a separate, stricter standard covering things like sensitive or low-quality content, and material that falls under them stays up for your followers while being kept out of Explore and the Reels feed. Account Status is where this shows up.

If a specific decision is listed there, appeal it from that screen. Meta operates a strike system with a published ladder — one strike is a warning, two through six bring time-limited feature restrictions, and all strikes expire after a year. Knowing where you sit on that ladder is more useful than guessing.

5. The metric changed underneath you

Views replaced plays and impressions as the primary metric across Reels, Live, photos, carousels and Stories, with the API cutover completing on 21 April 2025. Views counts repeats — one person watching your Reel three times is three views. Reach still counts unique accounts.

The consequence people miss: any benchmark you wrote down before August 2024 is measuring something different from what your Insights screen shows today. If your comparison baseline is a screenshot from 2023, part of the "drop" you are looking at is a change of definition rather than a change in performance. Fix the baseline before you diagnose the account. Our guide to what a good Instagram engagement rate looks like walks through which denominators are comparable to which.

6. The Reel simply did not hold attention

This is the cause nobody wants and it is still the most common one over a long enough window. Watch time is one of the three signals that matter most for Reels distribution, and Instagram measures it in seconds watched as well as percentage completed. A Reel that loses people at second two never accumulates the watch time to earn a second wave of distribution, so its view count stalls early and stays there.

If both follower and non-follower views dropped together, this is your answer, and no settings change will touch it. What to do about it is a longer conversation, which is why it has its own guide: how to get more views on Instagram Reels.

What a shadowban actually is, and is not

The word "shadowban" describes something that does not work the way people think. There is no secret throttle applied to your account that suppresses posts to your own followers.

What genuinely exists is loss of recommendation eligibility to non-followers, and Instagram tells you about it rather than hiding it. That is what Account Status is for. It is a real mechanism with a real appeal path, which is why the diagnostic above sends you there rather than to a third-party "shadowban tester". If you want the full version of that argument, we cover it in how to check for an Instagram shadowban.

Fixes that do not work, however often you read them

Every one of these appears near the top of search results for this problem. Every one of them is either dead or has been directly contradicted by Instagram. Doing them costs you time you could spend on the actual cause.

What changed in 2025 and 2026

If your mental model of Reels distribution was formed before last year, four things have moved and all four affect this diagnosis.

• April 2025: the API cutover completed and Views became the primary metric everywhere, replacing plays and impressions. Views counts repeats.
• December 2025: hashtags capped at five per post, caption and comments combined. Instagram's own framing is that fewer, more targeted hashtags perform better than many generic ones.
• 30 April 2026: the original-content rule expanded beyond Reels to photos and carousels. Repeat reposters lose recommendation eligibility across the app.
• 6–7 May 2026: an overnight sweep removed millions of bot, spam and inactive accounts. Typical small and mid-sized creators lost 2–5% of their follower count. If your views dropped in that window, check whether your follower count moved too — that is a different problem, covered in why your Instagram followers dropped overnight.

Worth noting on the May sweep: the celebrity loss figures reported in the press contradict each other badly across outlets. Treat any specific number you see quoted for a large account as unverified.

Where buying views helps, and where it cannot

We sell views, so here is the part that is against our interest and true. Buying views changes one thing: the number displayed on the Reel, and therefore what a person who lands on your profile assumes about it. That is social proof, and it is real — a Reel showing 200 views and a Reel showing 40,000 get read differently before anyone presses play.

What it does not do is restore recommendation eligibility. Instagram's ranking systems read rates derived from real viewing behavior — seconds watched, likes per reach, sends per reach. A purchased view count moves the numerator on a vanity metric without changing any of those rates. So:

• If your cause is the originality rule, buying views fixes nothing. No purchased number makes a repost eligible for recommendation again. Only weeks of original posting does.
• If your cause is engagement bait or a guidelines decision, buying views fixes nothing. Rewrite the caption or appeal the decision.
• If your cause is a watermark, buying views fixes nothing. Re-export the file.
• If your cause is that the Reel did not hold attention, buying views does not fix the Reel — though it does change how the next visitor reads your grid while you work on the content.
• If what you need is for a launch, a pitch deck or a brand DM to not show a profile of 200-view Reels, that is exactly what it is for.

If that is your situation, Reels views start at $4.99 for 20,000 and every view order completes within 1 to 12 hours, running at a steady rate rather than landing in one burst. General Instagram views work the same way. Views carry a completion guarantee rather than a refill, since a view once counted does not drop. We never ask for your password, and your profile has to stay public while the order runs. If you are weighing views against likes or followers, which to buy first is the more useful page.

A 30-day recovery sequence

If the diagnostic pointed at lost recommendation eligibility, this is the order to work in. It is deliberately unglamorous.

1. 01Day 0. Open Account Status. Appeal anything appealable. Screenshot it so you have a record of the date.
2. 02Day 0. Audit your last 30 posts. Flag anything that is someone else's work, carries a platform watermark, or contains an instruction to comment, tag or share. Remove or edit those.
3. 03Days 1–7. Post only original material. Do not change your posting frequency at the same time — changing two variables at once means you learn nothing from either.
4. 04Days 7–14. Track the follower versus non-follower split on every Reel. Recovery shows up as the non-follower line lifting first, usually gradually.
5. 05Days 14–30. If non-follower views are still flat and Account Status shows nothing, the problem is the content rather than eligibility. Move to watch time: seconds held, not completion percentage.
6. 06Throughout. Do not buy engagement to mask a diagnosis in progress. It will not change what the ranking systems read, and it makes your own numbers harder to interpret while you are trying to read them.

One more piece of context, because it prevents a lot of unnecessary panic. Instagram got harder for everyone: median engagement across formats was 0.45% per post by follower in Q2 2026, down roughly 24% year on year, and annual follower growth for accounts in the 1–5K range fell from 38% in 2024 to 22% in 2025. A gradual decline is the platform. A sudden cliff is a cause you can find.