Short answer: something on your side changed, and it's usually one of six things. A platform-wide shift is possible but rare, and it's the explanation people jump to first precisely because it's the one that isn't their fault.
Worth saying plainly: platforms do not single out individual small accounts. There is no penalty box. Whatever happened, it happened because of how the system read your recent posts against your recent audience.
The six things it usually is
1. You broke your cadence. A gap of two or three weeks is the most common cause by a distance. Recent activity is part of how you get distributed; come back after a silence and the first few posts go out cold while the system re-learns who you reach. This looks exactly like being punished. It's just a cold start.
2. You changed format. Photos to Reels, Reels to carousels, long captions to short. Each format is distributed differently and against different competition. A drop right after a format switch is a format result, not a reach result — and it often recovers once you've posted enough of the new thing to be measured properly.
3. Your first two seconds changed. If video reach fell but everything else held, look at your openings. A hook that starts with a logo animation, a slow pan, or someone saying "hey guys" bleeds viewers before the system has any reason to keep showing it. Retention drives reach far more than posting time does.
4. You saturated your followers. If you doubled your posting rate, each post reaches a smaller share of the same people. Total reach may be flat or up while per-post reach halves. This looks like a collapse in the per-post number and is often a non-problem — check the total before panicking.
5. Seasonality. School holidays, a heatwave, a major sporting event, the week between Christmas and New Year. Local businesses feel these hard, and they show up in reach before they show up in the till.
6. You started selling. A run of promotional posts tends to underperform a run of useful ones, because fewer people watch them through or save them. That's an audience response, not a punishment — and it's recoverable in a week.
How to actually diagnose it
Comparing this week to last week tells you almost nothing. Do this instead:
- Compare like with like. Reels against Reels, over 30 days, not 7. Single posts are noisy enough to prove any theory you like.
- Look at reach among non-followers specifically. If that held and follower reach fell, your existing audience has gone quiet on you. If non-follower reach fell, distribution changed. These are different problems with different fixes.
- Check saves and shares, not likes. They move first and they're what the system reads as "worth showing to more people." A drop here usually precedes a reach drop by a week or two.
- Find the last post that did well and the first that didn't. Something is different between them. It's usually visible in ten seconds once you put them side by side.
When it really is the platform
Occasionally it is. The signs are specific: it happens overnight rather than gradually, it hits every account you know rather than just yours, and it's being discussed openly by people who track this for a living.
Even then the response is the same, which is the frustrating and useful part: keep posting the things that got saved and shared. Distribution rules change. What gets someone to send a post to a friend doesn't.
What not to do
Don't delete the underperformers. You lose the engagement they did earn and learn nothing.
Don't panic-post daily. Doubling output into a soft patch usually deepens it — see saturation above.
Don't buy followers or engagement. You end up with an audience that never converts and a signal that tells the system to show you to more people like them.
Don't rebuild your whole approach off two weeks of data. Two weeks is noise. Sixty days is a signal.
The boring conclusion
Consistency recovers reach more reliably than any tactic. Nearly every drop we're asked to diagnose turns out to be a gap in posting, a format change, or a run of promotional content — all three of which fix themselves within a few weeks of going back to posting useful things on a schedule.
That is not a satisfying answer when your numbers just halved. It is, in our experience, almost always the right one.
Ready to stop posting at midnight?
Book a 30-minute discovery call. We'll tell you straight whether a content batch makes sense for your business — and which tier fits.