'Part 2 tomorrow' is how you halve your reach
The series mechanic that built half the channels of 2023 has flipped. Why follow-bait videos now underperform self-contained ones, and the one exception.
A creator I know — niche fitness, around 200k followers — went from 80-150k views per video to 20-40k in about six weeks last year. Nothing in their content changed. The thing that changed was that they started running a series. Every video ended on "Part 2 coming tomorrow." Every single one.
When they stopped doing that, views went back up within ten days. They didn't change the content. They just made each video resolve.
This isn't a one-off. I've seen the same shape on three other channels in the last six months. The "part 2 cliffhanger" mechanic that built half the creator economy in 2023 has flipped.
Why it worked in 2023
When the algorithm was newer, "follow for part 2" exploited something the recommender hadn't fully accounted for: the follow signal was worth a lot, and a video that converted a watch into a follow was treated as high-value. So you could trade completion rate (people leaving early to follow) for follow signal.
This is a coherent trade if the algorithm is mostly optimizing for graph growth. In 2023 on most platforms, it was.
What changed
The dominant signals in 2025-2026 are not completion or follow. They're replays and shares. From what I can measure from the outside, shares + saves combined seem to matter more than anything else.
A "part 2 tomorrow" video almost by definition can't be shared meaningfully — "hey, check out this video that doesn't actually say anything yet" isn't a thing people send each other. And it can't be replayed because the payoff is supposed to come in a video you haven't made yet.
So the trade is now: you give up the share signal (big) and the replay signal (big) for the follow signal (smaller, and only triggered by a fraction of viewers). Net negative.
The one exception
Series videos still work when each one resolves.
The good version: "Part 1 of why I think the calorie tracking thing is broken — today: why the math is wrong." Then tomorrow: "Part 2: what works instead." Each video is shareable because each makes a complete argument. The series tag is a hint to come back, not a withholding of value.
The bad version of the same series: "Calorie tracking is broken. Tomorrow I'll tell you why." That's not a series. That's a tease.
If you can't write a one-line takeaway for the video before you make it, the video is a tease. Teases die.
What to do instead
If you've been running a tease-series, the fix is brutal but mechanical.
Rewrite each video so it has a complete argument by the time it ends. Drop "part 2 tomorrow" from the script entirely. Put any next-video pointers in the description or end card, never in the spoken script. Number the series in the title or first overlay so binge-viewers can find the rest — but don't condition new viewers on having seen the previous video.
This is harder than it sounds, because once you've started running a series your next-video hook usually depends on the previous one. You have to break that habit. Write each video as if it's the only one that viewer will ever see — because for about 90% of them, it is.
Then, if they like it, they'll find the others on their own. The algorithm helps them.