How to Grow on YouTube by Tracking What Works

TL;DR

Consistent uploading helps YouTube growth only when each upload produces data that guides the next experiment. Across 90 days and 12 weekly videos, tracking click-through rate, average view duration, titles, and thumbnails revealed that documented experiences with visible proof gained more traction than generic advice, while improved packaging raised click-through rates from about 2% to 4–6%.

Transcript

You know how they say that's the secret to growing on YouTube is just to be consistent, like upload every week, show up, put in the work, and eventually the algorithm is supposed to reward you for it. And look, I'll be honest with you. I believed that for a long time, too, because it makes sense on paper. The more you post, the more chances the alg... Read More

Key Insights

• Consistency is a data-collection method, not evidence that YouTube will reward effort. Weekly publishing gave the creator repeated opportunities to compare results, but improvement began only after each upload became part of a measured experiment rather than another completed task.
• The first 30 days showed that generic advice was not enough to attract strong performance. Reviewing the videos together revealed differences in click-through rate and average view duration that were harder to recognize when examining uploads individually and immediately moving to the next production.
• A stronger video combined a recognizable person in its thumbnail with an exploratory format. The creator believed Ali Abdaal's presence helped generate clicks, while the higher view duration suggested that watching someone investigate and discover something in real time also sustained interest.
• Thumbnail testing raised the channel's click-through rates from about 2% during the first month to 4–6% during the third month. This improvement demonstrated that repeated publishing can support progress when each group of videos tests a specific packaging question.
• The best-performing pattern was documenting a real experience and displaying proof on screen. Across the 12-video experiment, these videos gained more traction than uploads built around generic advice, suggesting that an observable process can provide stronger viewer value than unsupported recommendations.
• Click-through rate is only one factor connected to video reach. The creator notes that YouTube also considers competing videos for the same audience, packaging, titles, and view duration, so a higher click-through rate does not automatically guarantee more views.
• A simple tracking page can make performance patterns easier to detect. The creator uses Notion to display each video's title, thumbnail, click-through rate, and average view duration together instead of navigating the larger collection of metrics presented within YouTube Studio.
• The proposed growth system has three connected parts: consistent uploads generate data, focused tracking turns that data into clarity, and title and thumbnail tools help convert findings into action. Because audiences evolve, the testing and measurement cycle must continue rather than produce a permanent formula.
• More videos from this channel:
  • How to Write YouTube Titles That Earn More Clicks
  • How Does YouTube Decide Which Videos to Push?
  • How to Grow on YouTube by Repeating What Works
• More from this channel: Kevin Chee

Questions & Answers

Q: Does uploading consistently make YouTube recommend a channel?

Uploading consistently did not produce any clear indication that YouTube favored the channel simply because more work was being done. Its practical value was generating a larger set of results to study. The experiment improved only when the creator compared uploads, identified performance differences, formed specific hypotheses, and changed future titles, thumbnails, or presentation formats based on those findings.

Q: How should creators use consistency to grow on YouTube?

Creators should treat consistency as a way to collect comparable data rather than as a promise of algorithmic rewards. Each upload should test something intentional, such as a thumbnail approach, title, topic, or presentation style. After publishing enough videos to reveal patterns, creators can identify what attracted clicks or retained attention and apply those lessons to the next experiment.

Q: Which YouTube metrics should creators track during an upload experiment?

The creator recommends tracking a focused set of information for every upload: the title, thumbnail, click-through rate, and average view duration. Arranging these details together makes comparisons straightforward and helps reveal why one video may outperform another. The experiment also shows that click-through rate should not be interpreted alone because viewing duration and competing videos matter too.

Q: What happened during the 90-day YouTube consistency test?

The creator uploaded one new video every week for 90 days, completing 12 videos. Generic advice underperformed during the first month, while exploratory videos based on genuine experiences showed more promise. After deliberately testing thumbnail approaches, click-through rates improved from about 2% initially to 4–6% in the third month, although total views remained limited by likely low viewing duration.

Q: Why did documented experiments outperform generic advice?

Videos documenting genuine experiences gave viewers a process to follow and included proof on screen. The creator observed that these uploads received the most traction among the 12 videos, while generic advice was less effective. The likely distinction identified in the experiment was that viewers could watch something being explored and verified instead of merely hearing recommendations about what they should do.

Q: How can creators improve YouTube click-through rates?

The experiment suggests improving click-through rate through deliberate, repeated testing of titles and thumbnails. During the second month, one of four weekly videos achieved a much higher rate and became that month's strongest performer. The creator then tried to replicate its packaging characteristics, eventually moving all third-month click-through rates into the 4–6% range from about 2% initially.

Q: Why does a higher click-through rate not always produce more views?

A higher click-through rate does not guarantee more views because YouTube evaluates several connected signals. According to the creator, the platform compares a video's rate with other videos competing for the same audience and also considers packaging, titles, and view duration. In this experiment, click-through rates improved substantially, but views remained relatively low, most likely because viewing duration was still weak.

Q: How can creators organize results from a YouTube growth experiment?

The creator organizes results in a simple Notion page containing every video's title, thumbnail, click-through rate, and average view duration. Displaying these elements together makes cross-video comparison easier than reviewing each upload separately in YouTube Studio. The purpose is not to capture every available metric, but to create enough visibility to recognize patterns and decide what to test next.

Summary & Key Takeaways

• The creator published one new video each week for 90 days, producing 12 videos while tracking their performance. The experiment showed that uploading regularly did not earn visible algorithmic preference by itself. Consistency became valuable when it supplied enough comparable data to identify patterns, test hypotheses, and make deliberate improvements.

• During the first month, generic advice videos generally underperformed. A stronger video combined higher click-through rate and view duration with an exploratory format. Its thumbnail included Ali Abdaal, but the creator concluded that its process-driven presentation also mattered because a familiar face alone could not explain the stronger viewing duration.

• The creator improved click-through rates from about 2% in the first month to 4–6% during the third month by testing and comparing packaging. The proposed system combines consistent publishing, a Notion page containing selected metrics, and title and thumbnail brainstorming with vidIQ, while recognizing that audience preferences continue to evolve.