End Screens Can Be Hidden: Measure Clicks Before Blaming the Button

YouTube’s end-screen element click rate is clicks divided by element impressions. It measures the response to elements that were displayed, not the share of viewers who started the video or reached its ending.

Interpret the rate as the final stage of a sequence: check how much of the audience remains when the end screen begins, confirm that elements generated impressions, then assess clicks and click rate. Viewers can hide the overlays, and end screens are unavailable in some viewing contexts, so a disappointing rate alone does not prove that the button, thumbnail or offer is the first problem.

Find the report and identify its denominator

For an individual video, open YouTube Studio, select Content, choose the video, open Analytics, select Engagement and scroll to “End screen element click rate.” At channel level, the relevant cards include “Top end screen element types” and “Top videos by end screen.”

The word “element” defines the denominator. YouTube’s reporting metric definitions record one impression for each end-screen element displayed, count clicks on those elements and calculate click rate as clicks divided by element impressions. They also define end screens as elements shown during the final five to 20 seconds of a video.

A playback that displays several elements can therefore produce several impressions. The denominator is neither unique viewers nor everyone who reached the ending. Do not confuse this measurement with the impressions click-through rate in Reach analytics, which concerns eligible thumbnail impressions that led to views.

Start with the audience that reaches the end screen

Use the audience-retention curve as a proxy for end-screen reach. Inspect the percentage still watching at the timestamp when the first element begins. This estimates how much of the audience arrived at the end-screen window; it does not establish that every remaining viewer saw every element.

YouTube’s audience-retention guidance places the report at video level, identifies dips as moments when viewers skipped or stopped watching and notes that retention data typically takes one to two days to process. Allow that processing time before treating an early curve as settled.

If retention drops sharply before the elements appear, the first weakness is audience arrival rather than element response. Examine whether a repeated conclusion, slow recap or detached outro is losing viewers before the recommendation. Moving the spoken transition earlier may help, provided it remains part of the promised content rather than interrupting the payoff.

If retention stays comparatively stable into the end-screen window, continue to element impressions. A new layout cannot recover clicks from viewers who left before an element could appear.

Separate display opportunities from creative performance

When viewers appear to reach the ending but element impressions remain scarce, check whether the elements were eligible, scheduled and displayed. Review their start times in the editor and compare videos with similar formats and end-screen timing.

YouTube’s end-screen requirements list videos designated Made for Kids, mobile web except iPad mobile web, the YouTube Music and YouTube Kids apps, Flash videos and 360 videos as contexts where viewers cannot see end screens. The same page notes that an end screen may not always appear, or may appear differently, because of performance, viewer behaviour, device and context.

Viewers also have direct control. The official Hide and Show explanation states that Hide dismisses all end screens for the current video and Show restores them. It also advises creators to integrate end screens into the video experience and prevent overlays from covering crucial visuals.

A low impression count does not establish that viewers selected Hide. The available end-screen metrics include impressions, clicks and click rate, but no hide count; eligibility, timing, device and context can also affect display. The defensible conclusion is only that fewer element displays were recorded.

Use clicks and click rate to assess the offer

Once viewers are reaching the ending and elements are generating impressions, clicks show the absolute amount of onward action while click rate adjusts that result for the number of displayed elements. Read both: more clicks can accompany a lower rate when the impression denominator is much larger, while a high rate based on few impressions may be too unstable to justify a major edit.

If impressions are substantial but clicks remain weak, the problem has moved downstream. Assess whether the promoted video or playlist is a natural continuation, whether the spoken call to action gives a concrete reason to continue and whether several elements are competing for attention. If an overlay covers the conclusion, viewers may hide it to finish watching instead of selecting it.

When practical, change one meaningful variable at a time. Keep the ending structure while changing the destination, or keep the destination while revising the spoken transition. Compare videos with similar audiences, formats and end-screen timing so differences in audience arrival are not mistaken for differences caused by the element.

Use a comparable baseline, not a universal target

The official documentation does not define a universal click rate separating a good end screen from a bad one. Element type, viewer intent, video format, traffic mix, timing and destination relevance all affect the comparison. A channel-specific baseline drawn from similar videos is more informative than an unsupported industry percentage.

1. Check retention: note the audience remaining when the first element begins and look for a preceding dip.
2. Check impressions: confirm that the elements generated enough recorded displays to interpret.
3. Check clicks: measure the absolute amount of onward action.
4. Check click rate: compare clicks per element impression within a genuinely similar group of videos.

Match the correction to the first weak stage. Low arrival points to the ending; scarce impressions call for timing, eligibility and display checks; healthy impressions with weak clicks point to the destination, transition or layout. Measure the full sequence before blaming the button.

Also read:

• The Era of "Premium" Streaming Is Coming to an End. And Viewers Aren’t to Blame.
• YouTube’s Biggest Thumbnail & Title Study Ever: 323,000 Videos, 62.6 Billion Views, and the Brutal Truth About What Actually Gets Clicks
• Top 5 Reasons Your Company needs a Blog and How to Measure Its Performance