It is usually not the algorithm

Before blaming distribution, there are three things worth checking, and they account for most clips that go nowhere.

01

What is the first thing to check?

Whether the clip reads with the sound off. A large share of feed viewing is silent, and a clip whose meaning lives entirely in the audio is a clip most people scroll past without ever knowing what it was about. Words rendered into the picture fix that, and it is the cheapest thing on this list to change.

02

Work down the list in order

Only after all five is the account itself worth suspecting, and even then variance explains most quiet fortnights.

CheckIf it is wrongFix
Retention in the first two secondsSteepest drop thereStart on the point
Words readable at arm's lengthType too smallFewer words, larger
Words clear of the interface bandCovered by app controlsMove them up
Sound level against other clipsNoticeably quieterRecord closer
Whether the moment stands aloneNeeds the surrounding ten minutesChoose differently

03

Checking the opening

Watch the first second of your recent clips. Does something happen immediately, or does the clip start with a greeting, a pause or a blank frame? The first moment decides whether most viewers stay, and a slow start is one of the most common reasons clips go nowhere.

Start on the point, a reaction or a clear question.

04

Checking whether it stands alone

Show a clip to someone who does not know the full conversation. If they cannot explain what it was about, the clip depends on missing context. Clips that stand alone travel much further.

Add a line of setup or choose a different moment.

05

Checking the words on screen

Many viewers watch without sound. If your clip has no words on screen, or words too small to read, those viewers have no reason to stay. Clear, readable text is one of the cheapest improvements available.

Test on a phone with the sound off.

06

Checking consistency

Posting occasionally and irregularly makes it hard to build an audience. A steady schedule, even a modest one, helps viewers and feeds recognise your account.

Aim for a rhythm you can keep.

07

Giving it enough time

Early clips on a new account often get few views while the account builds history. Judge progress over dozens of clips and several months, not a handful.

Track the median rather than individual clips.

08

Learning from the best performer

Look at your best clip so far and ask what made it different. Repeat that element deliberately.

Small wins point the way forward.

Write it down and check the next five clips against it.

09

Check the cheap things first

  • Watch it on mute. If you cannot follow it, neither can most of the feed.
  • Check the first second. A clip that opens on throat-clearing has already lost.
  • Ask if it stands alone. Needing the original video is the most common quiet killer.

Cut one video with the words in the picture and compare.

Questions people ask

Is posting more the answer?
Only after the three checks above. More of something that does not work is just more.
Does clip quality matter?
Less than the first second does, though a watermark on a reposted clip does read as recycled.
Should I change platforms?
Rarely the problem. The same clip usually performs similarly across all three.
How many clips before I can tell?
Enough that one bad week does not decide it. A handful is not a sample.
Do subtitles really change this much?
In a silent feed they are the difference between a clip being understood and being scrolled.

Read next

01Watch time or viewsGuide02Subtitles cut off on phonesGuide03Views stop at two hundredGuide