YouTube engagement rate is a calculated ratio, not one universal number with one universal formula. Before you compare two videos, decide which interactions you are including and what you are dividing them by.
A view-based rate using likes and comments will produce a different result from a subscriber-based rate that also includes shares. Both can describe something, but they cannot be compared honestly without explaining the method.
A simple view-based formula
For an internal review, one straightforward method is: (likes + comments + shares) ÷ views × 100. Use figures for the same video and reporting period, and include only metrics you actually have.
If you do not have reliable share data, use a clearly labeled likes-and-comments rate instead. Do not silently mix a full interaction total for one video with a narrower total for another.
This is a chosen analysis formula, not a claim that YouTube defines every creator’s engagement rate this way. YouTube’s own engagement reporting includes metrics such as watch time, average view duration, audience retention and likes versus dislikes.
A worked example with fictional numbers
Imagine a video has 2,000 views, 80 likes, 15 comments and five shares in the period you are reviewing. The included interactions add up to 100. Dividing 100 by 2,000 and multiplying by 100 gives a 5% view-based interaction rate.
Those numbers are illustrative, not a performance claim about GetPlaysNow or a customer campaign. The result means there were five counted interactions for every 100 counted views under that formula. It does not mean exactly 5% of unique people interacted, because one person may produce more than one type of interaction.
If the view count is zero, do not calculate the percentage. Report that the rate cannot be computed for that period instead of inventing a result.
Why the denominator matters
A subscriber-based rate asks a different question from a view-based rate. Subscriber count includes people who may not have watched that particular video. Unique viewers, meanwhile, is an audience estimate rather than a raw playback count.
Keep your chosen denominator visible in the spreadsheet heading. A label such as interactions divided by views is much clearer than a column simply called engagement.
Public-view definitions can also change over time. YouTube’s current counting guidance distinguishes public views from engaged viewing measures. If your comparison spans a measurement change, note that limitation rather than treating every historical percentage as directly comparable.
Compare similar videos, not unrelated channels
A detailed tutorial, a short announcement and a music performance invite different behavior. One may solve a problem without generating many comments. Another may ask an explicit question and produce a discussion.
Compare videos of similar format, age and purpose. Separate Shorts from long-form content where the viewing behavior and reporting differ. Note unusually small samples: a few extra interactions can move a percentage dramatically when there are very few views.
There is no useful universal percentage that proves every kind of video succeeded. Set a baseline from your own comparable uploads and look for changes you can explain.
Turn the calculation into a decision
If viewers are watching but rarely responding, try a relevant, specific question rather than several generic requests. If the opening loses people, improve that section before focusing on likes. Read the retention report alongside your calculated rate.
When comparing paid options, the YouTube likes service information and 1,000-like product listing show the relevant offering. Purchased activity is not evidence of authentic enthusiasm, and artificial engagement can be removed or create policy risks. Do not use it to misrepresent audience research or sponsor results.
What should go in the final report?
Include the video URL, date range, raw inputs, formula and a short note about the video’s goal. Add what you plan to test next. That gives a collaborator enough information to understand the result without relying on a percentage taken out of context.
Related reading
YouTube Impressions vs Views: Find the Real Bottleneck
