Your last interview attracted more views than your latest tutorial. That does not necessarily mean you should stop making tutorials.
The interview may be older, cover a newsworthy guest or receive traffic from a large external audience. A useful comparison accounts for those differences before turning a result into a channel strategy.
YouTube's Advanced Mode lets you compare videos, groups and time periods, adjust reports and export data. The purpose is to answer a defined question more carefully. YouTube's documentation describes the available controls.
Write the decision before opening the report
Start with a question you can act on: “Should the next production block prioritize troubleshooting tutorials or interviews?” Then write down what you need to know.
For a hypothetical software education channel, the decision might involve sustained viewing, relevant subscribers and production effort. A total view count covers only part of that decision.
If you are still learning the basic reports, use the weekly YouTube analytics routine before building more elaborate comparisons.
Build groups with clear rules
Create a tutorial group and an interview group using explicit membership criteria. Keep Shorts, live streams and ordinary uploaded videos separate when their formats and metric definitions make a pooled result hard to interpret.
Do not put a weak tutorial into the interview group because it resembles a conversation. Consistent classification makes the comparison worth repeating.
Add your own notes for factors the report cannot know: paid promotion, a guest's newsletter mention, a major news event or unusually high production costs. These notes help explain differences without inventing algorithmic reasons.
Compare the same part of each video's life
An older video has had longer to collect activity. YouTube documents comparisons over the first 24 hours, seven days or 28 days of a video's lifespan, alongside group and filter workflows. Choose a window that suits the question. Its Advanced Mode tips explain these options.
A tutorial that answers a persistent question may develop differently from a timely interview. Inspect both an early window and a later one if that distinction matters. State the window in your report so someone else can reproduce it.
Use this comparison plan:
| Comparison | What it may reveal | What to check next |
|---|---|---|
| Similar videos at the same age | Differences in early response | Topic, packaging and distribution |
| Groups split by traffic source | Dependence on external or search traffic | Whether the intended audience is represented |
| Similar-length videos | Viewing patterns within a format | The actual footage around departures |
| Production effort versus outcomes | A format's practical sustainability | Which work can be simplified |
Inspect individual videos behind the aggregate
A group total may be dominated by one exceptional upload. Review the individual rows before declaring the entire format successful.
If you export the data, calculate a typical result alongside the total—for example, a median across the selected videos. Label this as your own calculation, record the metric and keep the membership list. Do not present it as a separate native YouTube metric.
Small groups deserve cautious conclusions. Three videos about different topics cannot isolate the effect of a new presenter or editing style. They can suggest a question worth testing again.
Turn an observation into a testable change
Keep three fields separate:
- Observation: tutorials received more of their traffic from search in the selected window.
- Hypothesis: their specific problem statements may help viewers recognize the answer they need.
- Action: publish another tutorial with a clear question and a complete demonstration, then compare it with similar uploads.
The observation is visible in the report. The hypothesis needs further evidence. The action gives the team something concrete to do.
For packaging and retention questions, follow the YouTube performance diagnosis guide. For search intent, use the YouTube SEO guide.
Save the comparison with its assumptions
Save a reusable report where available and record the group, period, filters and metric definitions in your notes. Treat unavailable or restricted data as missing evidence rather than zero activity.
The final recommendation should name a production decision, the evidence supporting it and the factors that could change your mind. A carefully bounded conclusion is more useful than a confident story assembled from mismatched totals.