Did sharing your YouTube video on Reddit hurt its retention? Read the mix
A worked traffic-source example shows how average view duration can fall without a change in existing viewers. Decide where sharing actually helps your audience.
The quick answer
External visitors can change an overall viewing average, but a lower average alone does not show that sharing damaged recommendations. Separate traffic sources, compare the same time windows and check whether the community wanted the video before changing your promotion strategy.
In this guide
A creator discussing external traffic on Reddit noticed brief visits from creator-feedback communities and longer viewing from YouTube traffic. The worry was understandable: could asking for feedback be harming the video?
There are two questions here. Did the mix of viewers change the average? And did that change cause a recommendation outcome? The first is arithmetic. The second needs evidence that a single aggregate graph cannot supply.
A falling average can be a mixture effect
Consider an illustrative video with 100 viewers averaging four minutes. That is 400 minutes of watch time. Now add 100 visitors from a feedback thread, each averaging thirty seconds: another 50 minutes. The combined average is 450 minutes divided by 200 views, or 2 minutes 15 seconds.
Illustrative arithmetic, not measured channel data. The original group's average stays unchanged.
The original group has not become less engaged in this example. Its viewing time is unchanged. The overall average fell because the second group behaved differently. These are invented numbers for explaining a weighted average, not a report about Reddit traffic in general.
| Group | Views | Average view duration | Watch time |
|---|---|---|---|
| Original group | 100 | 4:00 | 400 minutes |
| Added external group | 100 | 0:30 | 50 minutes |
| Combined | 200 | 2:15 | 450 minutes |
Do not average 4:00 and 0:30 unless the groups contain equal numbers of views, as they do here. With unequal groups, weight each average by its view count.
Inspect the source before changing the video
YouTube's video-reach documentation describes traffic-source reports, including external sites and apps. Use the available source detail to distinguish a relevant community discussion from a general request for creators to inspect each other's work.
Record when and where you shared the link. Compare traffic and viewing data over matching periods, and separate sources where the reports allow it. Small groups can fluctuate sharply; a handful of short visits is a weak basis for a channel-wide rule.
Also distinguish average view duration from average percentage viewed. The arithmetic above concerns time watched per view of one video. Comparing different video lengths introduces another variable.
Share for a viewer need, not a view counter
Suppose you made a tutorial on repairing a particular bike brake. A community member asking about that exact repair has a reason to watch. A general “please rate my channel” thread has a different purpose: people may inspect your opening, leave feedback and move on.
Before sharing, read the community's rules, answer the question in the discussion where appropriate, and explain what the video adds. Do not flood communities with links or disguise self-promotion as an independent recommendation. Useful sharing starts with a match between the question and the video.
If the goal is production feedback, treat the responses as feedback. You may learn that a diagram is unreadable even if the reviewer does not finish the video. That is different from measuring demand for the topic.
What the numbers cannot settle
A before-and-after drop in views does not isolate the effect of an external post. Distribution may already have been changing. The topic may have limited demand, or the comparison may cover different hours and different audiences.
Keep sharing that helps the intended viewer, stop promotion that creates no useful audience connection, and record the reason. Avoid either blanket claim: that all external traffic damages a channel, or that external traffic can never be associated with poor viewing. The actionable question is whether this placement reached people who wanted this video.
Common questions
Should I delete a Reddit post if the video average falls?
An aggregate drop alone is not enough to decide. Check the community rules, the relevance of the placement and source-level data. Remove a post that is inappropriate or misleading, rather than treating deletion as an algorithm repair.
Can I calculate YouTube CTR by dividing all views by impressions?
No. External views and other views may not originate from counted thumbnail impressions. Use YouTube's reported impressions CTR and its definition, rather than combining incompatible totals.
Published by Vidfora, the product discussed in these guides. Examples are illustrative unless a source is named. Editorial approach · Suggest a correction