How to Compare Two YouTube Videos Fairly

Comparing your video with someone else's is one of the most common uses of public data, and one of the easiest to get wrong. The numbers are real. The trap is in what they leave out, and in how the arithmetic treats age and format.

What the tool compares

Paste 2 to 5 video links or IDs. The tool reads all of them in one YouTube Data API videos.list request, and Google's documentation gives a call to that method a quota cost of 1 unit, so adding videos does not cost more. For each video it shows:

• Views, likes and comments: public counts. The API documents them as the number of times the video has been viewed, the number of users who liked it and the number of comments.
• Engagement rate: likes plus comments, divided by views.
• Views per day: views divided by whole days since publishing, with a minimum of one day.
• Duration.
• The number of tags the API returned. It is a count, not the tags themselves.

The video that is strictly ahead on views, likes, comments, engagement rate or views per day is marked as the leader for that row. Duration and tag count are shown but never marked. Ties, and rows where all but one video hides the number, stay unmarked, because the tool would otherwise be guessing. Private and deleted videos cannot be read and are skipped, and the page says how many.

A worked example on three real videos

This is our example, not a benchmark. On 25 September 2026 at 15:53 UTC we fetched three public videos with one videos.list call and applied the tool's formulas. We picked them to show the traps: an old classic, a video from five days ago and a Short from the same channel.

Four things in that table are worth reading slowly.

Total views and views per day disagree. The 2009 video has about 28 times the views of the newest one, and the newest one averages about 44 times as many views a day. Neither ranking is wrong. They answer different questions, one about a lifetime and one about an average day.

The daily figure is only as good as the day count. The tool counts whole days, so a video published 5 days and 23 hours ago is treated as 5 days old. Using the exact elapsed time, 5.995 days, the newest video averages about 10.9 million views a day, roughly 17% below the 13.1 million shown. A whole-day rule flatters very new videos most.

An average hides the shape. The API returns totals, not daily counts, so we cannot see how the views were spread. Our reasoning, not a measurement, is that a lifetime average for a 17-year-old video blends its early years with everything since, and a five-day average for a new release blends a launch with what follows.

Format and age change together. The Short has 15.2 million views in 32 days, and the long video has 65.5 million in 5. Its views per day are about 3.6% of the long video's. They differ in length, age and audience at once, so no single cause can be assigned to the gap.

Traps to check before you draw a conclusion

1. 01

   Compare views per day, then check the age

   A very new video makes the daily figure jump around, and the whole-day rule leans it upward.

2. 02

   Match the format and length

   A 36-second Short and a 19-minute video attract different behaviour. A comparison across formats says little about which one is better.

3. 03

   Check for hidden numbers

   YouTube Help says a creator can choose whether viewers can see how many likes a video has. The tool treats a count it does not get back as missing, not as zero, and an engagement rate built on comments alone can look lower than it should.

4. 04

   Remember when the counting rule changed

   YouTube Help says that beginning 24 August 2026 views are counted the moment a video starts to play, across Shorts, long-form videos and live streams. We found no statement on whether earlier views were recounted, so treat a total that spans that date as mixed.

5. 05

   Do not read the tag count as quality

   In our example the video with 27 tags is not the one with the most views per day. The count tells you what was set, not what worked.

6. 06

   Compare on the same day

   YouTube Help says metrics can take a few hours to appear on a new video, and that YouTube may temporarily slow down, freeze or change a count to keep it accurate. Two readings taken days apart are not comparable.

What public numbers cannot show

YouTube Help describes watch time, impressions, impressions click-through rate and average view duration as metrics in YouTube Analytics, and it says the audience retention report is only available at the video level of YouTube Analytics. The public API response the tool reads carries none of them. If you want to know why one of your own videos held viewers better than another, that answer sits in your Studio, not in a side-by-side of public counts. Our Video Comparison shows how big the gap is, and the Channel Audit shows whether it is typical for that channel.

FAQ

Related reading