← Projects

Creator analysis

I wanted content ideas that came from watching what people actually respond to.

Inside the research

I built a system that studies creators’ videos, the words on screen, the opening, the cuts, and the comments underneath. It maps what happens moment by moment, so I can ask why an opening works and get back the source.

It also compares a post with that creator’s usual performance. A big view count alone doesn’t tell me much.

Video frames and opening sequences from Morilliu and Meshtimes, compared with each account’s sampled median performance
The build becomes the story. Frames from @morilliu and @meshtimes; analysis from my September 2026 working report.
Audience questions grouped from 40 selected comments, with counts explicitly limited to that sample
What people are asking for. These are selected comments I grouped for this research, rather than a demographic survey.

Ask about a moment

I can ask where a tutorial moves from instructions into proof, and get the exact timestamp, spoken words, and frame. The video stays connected to its evidence instead of becoming one long summary.

Public views and comments help me compare posts. They don’t tell me where viewers dropped off, and they don’t prove that an opening caused a video to perform.

Nine timestamped stages of a tutorial and a research query returning the exact moment the creator demonstrates proof
A query against a tutorial by @itsmariahbrunner, from the same report.

From research to an idea

The useful part is connecting an audience question to something I can make: a small tool, a tutorial, a story, or a new way to explain an idea. Each suggestion needs a first frame, something worth watching for, and an actual result for the viewer.

This is an ongoing experiment inside Lola OS. I’m still testing which recommendations deserve to become posts.

More about Lola OS
A proposed content concept connecting an audience question about technical language with an opening, first frame, editing sequence, and useful takeaway
A proposed concept, not a published result. Its source structure is credited to @shebuildsrobots. The report is a working document; I review what becomes content.