Strategist answers · AI niche · 50K to 500K
Our four-person AI Shorts channel is losing views at 145K subs; how do I reduce editing work when every automatic cut we've tried removes the pauses that make our experiment reveals land?
For 100K-200K AI niche channels; small team (2-5); Shorts-first; declining.
81% of this answer is quoted word for word from 6 YouTube strategists. Each quote links to its source. Last updated .
The 1of10 Podcast and 1of10 offer contrasting approaches to pauses for a four-person team making AI experiment Shorts:
“It scans your video, understands the different takes. For example, here take one, take two, take three, chooses the best, and then uh brings it all together. So, for example, this video was initially all the gray area was cut out, all silences, all repetitions, and then cuts it into a 1 minute 7 video. So, something that took hours now just takes a couple minutes.”
“Your voice and body language control how viewers feel. Use pacing and tone intentionally: Slower pacing → serious moments Higher energy → exciting moments Softer tone → calm or reflective moments Pauses → emphasis Monotone delivery kills storytelling. Emotion amplifies it.”
For the reveal timing in those Shorts, Ed Lawrence adds another editing perspective relevant to the pauses being lost:
“It's jarring. It's messy and it pulls viewers out. It makes you look bad. It wrecks your credibility. It feels like a glitch. So, only do it when the sentence come to a natural stop like that. See the difference? Now, I've also cut back on jump cuts about 1000 times. But, if you're unsure how to use a jump cut, just don't use it. Just cut on the same width like this.”
On the editing workload facing this four-person channel at 145K subscribers, Ed Lawrence describes a separate part of the process:
“It took 40 minutes of out-of-order footage, reordered it correctly, cut it up, and added basic text and graphics without making a mistake at a speed that is not possible for a human to do. I was a pro editor for five years, I am very fast at editing, but that would have taken me about an hour and it would have been a really boring, annoying job I would have hated.”
Alongside that example, Dabi offers a caveat about AI tools for a team trying to preserve its experiment reveals:
“Super excited and stay tuned, it’s actually just absurd what is being cooked. Taking a while precisely because of the taste aspect of things; even with code/architecture, I myself have a strong internal conviction on what I think “good” looks like. Many of the tools will be really freaking cool. They won’t be useful to everyone, but they will be super dope for a few YouTubers, and that specificity is more exciting to me.”
For the channel’s declining views, Leroy ter Braak and Paddy Galloway provide two further angles on reviewing the work:
“Execution - Lay out a plan to execute better than the competition already on YouTube Analysis - Watch your final product with hawk eyes and figure out what can still be cut out of the video without losing context Longevity - Ask yourself if this video will bring you one step closer to your long-term goal.”
“Unless you’re a massive channel, look at: - Views: what’s worked and what hasn’t - Retention curves: where do we lose? Everything else is noise at the beginning.”
Join Coeditor's invite-only beta waitlist to try rough cuts planned against 90 storytelling rules. Each keep and cut has a readable reason your team can challenge. Coeditor
Every quote is copied word for word, credited to the person who said it and linked to the original post or video. Spotted a misquote or a missing credit? Tell us and we'll fix it.