← All case studies

Why AI & Faceless YouTube Channels Die After a Viral Video

To start a successful channel fast, can you compromise on quality to focus on quantity, and then swap to quality when you get a viral hit/are monetized?

1.4 million viral hit Dead channel Failure Click to see By Kai Jörgsen Väss · Frame 117

To start a channel fast, can't you compromise on quality, use AI, focus on quantity and then swap over when you get a viral hit/are monetized?

I've had this question come up quite a lot lately, and mostly it's about using VidRush or some low quality editor to make many videos, fast. Even if the IDEAS themselves and the positioning gap is GREAT, it's almost always without fail…. The wrong play.

Let me tell you why….

Normally it's good to start testing 5 quality videos but they can take a long time to create and you might lose your "first mover advantage" (being able to beat everyone else to an opportunity.). Which is true… speed IS important. And some niches will reward slop spam before it dies. But almost ALL successful channels on YouTube will end up testing with some sense of quality to start… and have that improve in the first 5-10 videos they launch. I use "quality" here broadly but it really comes down to your ability to have an amazing script, great VO, and effective editing applied to a fantastic idea.

And all it takes is ONE idea to go boom. And if that video is high quality? Your audience will love it, subscribe, and be waiting for me. After you've read this, go read the 01 Case Study on my Gravipull channel to see what you do in that situation…. But what if you've used lower quality or AI where you could have 5 videos before a viral hit or 20…. Why can't this channel also succeed?

See, it's unlikely once you have that 1 million view video that you'll have a team ready to continue momentum on a Vidrush/AI channel that's able to make content that the return audience expects, especially if you're going docs/mini docs/animated….

I looks good on the outside but...

Imagine you do 10 videos with Vidrush, get the views, and want to convert to a human team to avoid any inauthentic claims and reward your audience for watching so far. But, you can't replicate the style that everyone has come to expect. Yes, even if it's low quality and simply made. They clicked and watched that video for a reason. The market determines everything. However, that might be the case for some channels…. It's actually the icing on a shit cake. Let's break this down further:

Ultimately, you're way LESS likely to have a core returning audience due to the heavy use of AI, or if it's just an "automation" channel, you're way less likely to have a returning audience because your scripts suck and you don't deliver on an amazing idea's packaging or curiosity that the viewer expected when clicking… So what happens? Your average view duration (AVD) is low, and in the event it isn't and people feel like "hey, this video was great.. Cool!", they will still WAY less likely to subscribe or ever click on your videos again. It was good enough for a watch… one watch.

After this viral hit, sure, you'd be monetized… but now you're effectively starting from zero trying to get your current viewers to now buy into a different style of video from one to the next. It'd be quite jarring. Or you're trying to get new viewers all over again.

But I have receipts from channels out there (not from here at 117 as we do not believe in shortcuts or clickbait screenshots of one video doing the up up ups in the graphs)… actually, we have a LOT of receipts but let's focus on this channel for today:

The viewers aren't happy.

That's a HUGE drop off even for a Part 2

See attached image from a one million + view video…. an AI voice, very clearly AI production.

He had a hit with AI, everyone was tilted in the comments that it wasn't high quality, didn't subscribe, and aren't going to be returning viewers. Because of this, he lost all momentum entirely. However, with the million view video he progressed to BETTER visuals, less obvious AI, and a HUMAN VO which you can see for Part 2 of this video. Doing a Part 2? Very frequently a terrible decision anyway, but it's got 100k views only in comparison. And look at the rest of the videos around it.

Yeah, his ideation and strategy isn't the best, but if you can't convert a 1 million+ view video into a LARGE conversion of subs and returning viewers? It was a quick payday and meant nothing else.

Using too low quality for a new channel means this is a highly probable outcome. Regardless of what that looks like.

It's just better to have your quality team set up and in place from the get go... because again: animators and writers and VOs aren't going to just sit around waiting for you to summon them on 2 week notice. They will get other jobs. It only works if you have a literal team of people on full time retainers working for you anyway.

Now you can compromise a bit on quality for speed and increase it as you go (See @Crayon Capital for example of this) but it's still gotta convert people initially into a returning viewerbase.

Of course it CAN work, not saying it can't... it just feels like splitting focus between A and B. There are still risks involved with investing in fewer videos because Youtube isn't a magic game and we aren't Gandalfs - no one is. You have to assess the best data and make the best content.

As always, a game of risks and balance! But that's why we have a "Build With You" option here at Frame 117 to give you the best possible chance of success and making that risk low, and that balance ever in your favour.

Want this for your channel?

Join the Frame 117 community, or book 1-on-1 consulting with Kai & Andrew.

See the offers →