Why Is Youtube Shorts Algorithm So Bad?

YouTube Shorts gets over 50 billion daily views and is one of the main ways people consume video content every year.

However, many creators and viewers have been complaining that the new YouTube Shorts algorithm seems broken and has become really bad at recommending relevant Shorts.

And to be honest, there is some truth to this.

In this post, I’ll talk about why the Youtube algorithm is so bad and the problems it has caused for both viewers and creators.

The Issues With The Shorts Algorithm For Viewers


The main problem for users and viewers with the algorithm is that it started showing a bunch of useless stuff people have 0 interest in.

This is because Youtube split off the watch history and removed the bridge.

Shorts history is only used for shorts recommendations, and long form recommendations are based on only long form watch history.

So it’s only going to show you shorts similar to the ones you saw after the new update went live.

And if you interact with any of these videos like clicking on the dislike button or clicking not interested, you are still engaging with them and you’ll just see more of them.

⚡Also Read: Should I Trademark My Youtube Name

What You Can Do

Luckily, fixing this is pretty simple now that we know the recommendation system works by looking at your shorts watch history.

First up, STOP going through your shorts feed. Do not click it for the next couple of days.

Now, use the Youtube search to search for stuff you find interesting and then click on the shorts that are shown.

Go through a few and then do another search for something else you like.

Keep doing this for a few days but don’t go to the normal shorts feed.

You’ll soon start to see your feed recommendations will be similar to the stuff you searched in the past few days.

⚡Also Read: Can I close Youtube while processing?

The Issues With The Shorts Algorithm For Creators


The biggest issue for creators with the algorithm is views. The algorithm is so bad at this.

The shorts algorithm works by showing the new short you uploaded ONLY to what they call a “seed audience” which is a few hundred people, and see how the videos does.

If they engage very well with the video, the algorithm pushes the short into more feeds. But if they don’t your short will just stop getting views – forever.

And the crazy thing is the Seed Audience is not super targeted and are just people who this AI algorithm thinks MIGHT like it.

If your video lands on the feeds of people who are not really into your niche, your videos will never take off.

On top of all that, the issues the viewers are facing will also affect creators a lot.

Because there is no connection between shorts and long form video watch history, the algorithm won’t recommend long term videos based on the shorts people watch.

And this is the reason why none of your long form videos are taking off when you keep getting thousands of short views and likes.

The long form videos are just not recommended to them.

Because the long form algorithm doesn’t know the user likes your content.

What You Can Do

The only thing you can do is to make your video more engaging and post at the right times.

There are some great tips about that in this video:

We also have a couple of good guides on what to do if your shorts are not getting views here:

Unfortunately, there’s nothing much you can do about the other thing.

Is The Youtube Shorts Algorithm Broken?

The Youtube Shorts algorithm was broken but it isn’t the case anymore. All the issues we talked about are by design. And that is how it will be until they release a big update.

Why Youtube Shorts Not Going Viral?

The reason your shorts are not going viral is because they are not engaging enough. Your short needs to be like SUPER awesome and everyone should love it. And it has to be a trending topic.

About Author

Lito James is the founder of Massivepeak.com. He is an entrepreneur and marketing specialist who helps businesses to get more leads, subscribers, and customers. Massive Peak has been featured on G2, Cloudways, Sujanpatel, GetResponse, Renderforest, and many more. Follow on LinkedIn | Twitter

Leave a Comment