Audience Retention shows what percentage of a video viewers watch before leaving. YouTube provides a "Key moments" report that identifies four specific moment types in each video — each with a different diagnosis and a different recommended action. The report is accessed in YouTube Analytics at the video level only.
| Moment type | What it shows | What to do |
|---|---|---|
| Intro | % of audience still watching after the first 30 seconds | If low: adjust thumbnail/title or rework the first 30 seconds |
| Top moments | Moments where almost no one dropped off | Introduce this type of content earlier in the video |
| Spikes | Moments that were rewatched or shared | Investigate: engaging content — or confusing content? |
| Dips | Moments that were skipped or where viewers left | Review why audience interest dropped at that point |
Key takeaways:
- Per the Key moments page, the retention report is available at the video level only — not channel level. Requirements: video at least 60 seconds long and at least 100 views. Data takes 1–2 days to process.
- Per the official page, "Typical retention" (the gray line) is the average of your last 10 videos of similar length — not YouTube's global average. Performing above this line means outperforming your personal baseline.
- Per the official page, Spikes do not always signal great content — they may mean "the content wasn't clear and the audience had to rewatch a part."
- Per the official page, "Intro" shows what percentage of your audience is still watching after the first 30 seconds. Videos where 50% or more watch past 30 seconds are listed as "above typical intros."
- Per the official page, "if your top moments happen later in your video, consider introducing that content earlier — audience numbers tend to decrease as videos progress."
How to access and read the Audience Retention report
Per the official page, access: YouTube Studio → Content → select the video → Analytics → Overview or Engagement tab → Audience retention report.
"The shape of the audience retention graph can tell you which parts of your video are most and least interesting to viewers." Per the official page, you can click "See more" to compare your video against all YouTube videos of similar length — in addition to your own typical retention benchmark.
Intro — the first 30 seconds determine retention
Per the official page, Intro shows what percentage of your audience is still watching after the first 30 seconds. The official interpretations:
Interpreting the Intro metric per the official page:
- High percentage (positive): "The content in the first 30 seconds matched the viewer's expectation of the video's thumbnail and title" and "the content kept the audience interested."
- Low percentage — recommendation 1: "Consider changing your video thumbnail and title to better reflect your video content."
- Low percentage — recommendation 2: "Modify the first 30 seconds of your video and experiment with different styles to find one that will keep your audience engaged."
A low Intro rate has two distinct possible causes: either the thumbnail/title over-promised and the viewer left on discovering the content doesn't match — or the content itself failed to engage quickly enough. Diagnosing which one drives the appropriate fix.
Top moments — content where almost no one leaves
Per the official page, Top moments are "moments in your video where almost no one dropped off while watching." Actions per the official page:
- Find and study these moments — what do they have in common across your different videos?
- Per the official page, "if the top moments are occurring in a later portion of the video, consider introducing the compelling content earlier in the video — audience sizes typically decrease over the length of the video."
Spikes — two very different interpretations
⚠️ Spikes do not always mean great content
Per the official page, Spikes are moments in your video that were rewatched or shared. This "could mean your audience watched that part more than the rest of the video because it caught their interest" — or it could mean "your content wasn't clear and your audience had to rewatch a part." Look at the content at that timestamp to determine which applies: a funny or useful moment (positive signal) vs. a complex explanation or unclear transition (needs improvement).
Dips — diagnosing where interest drops
Per the official page, Dips highlight moments "that were either skipped or moments where viewers stopped watching your video completely." The official recommendation: "Review where dips are in your videos to better understand why the audience lost interest."
Common diagnosable causes of Dips:
- Long introductions delaying the actual content.
- Abrupt shifts in pace, tone, or subject matter.
- Mid-roll ads — per the official page, Dips at ad timestamps are expected and should not be the primary focus of optimization efforts.
- Padding or repetition after the core content has concluded.
Typical retention — the correct benchmark for comparison
Per the official page, Typical retention (shown as a gray line in the report) is "the engagement that the last 10 of your videos of a similar length have maintained." It is your personal benchmark — not a YouTube-wide average.
A video performing above the Typical retention line is outperforming your personal baseline — even if the absolute retention percentage appears lower than what you see from larger channels. Comparing against your own trend is the documented first step before comparing against industry benchmarks.
Retention Segments — a deeper audience read
Per the official page, the Segments report within Audience Retention allows comparison between:
- New viewers vs. returning viewers: If new viewers drop off much earlier, the intro may assume prior context they don't have — or the content doesn't match what brought new viewers in.
- Subscribers vs. non-subscribers: If subscribers complete a higher percentage, the content resonates with established fans. If non-subscribers complete more, the video has broader appeal than your usual content — a signal worth acting on.
- Organic traffic vs. paid traffic (for channels running ads): Compares retention behavior between viewers arriving from search/recommendations and viewers arriving from ads.
Frequently asked questions
Why doesn't the Audience Retention report appear for my video?
Per the official page, the retention report requires two conditions: the video must be at least 60 seconds long and must have accumulated at least 100 views. The data also takes 1–2 days to process after those thresholds are met — so a video that just crossed 100 views may not show the report immediately.
Is a 50% audience retention rate considered good on YouTube?
YouTube does not specify a universally "good" retention rate that applies across all content types and lengths. The documented benchmark is your own Typical retention — your last 10 videos of similar length. The official page notes that videos where 50% or more of the audience watches past the 30-second mark are listed as "above typical intros" — but this specifically refers to the Intro metric (first 30 seconds), not overall retention for the full video.
Can improving retention on an old video improve its ranking?
Retention on an already-uploaded video cannot be retroactively changed — it is measured from actual viewing behavior. What can be changed: the thumbnail and title (which may attract a better-matched audience whose viewing behavior naturally produces higher retention on new views), or the use of retention insights to improve the structure of future videos. Cards and end screens can also be added to direct viewers to more content after they've finished watching.
Do retention Dips at mid-roll ads hurt video ranking?
Per the official pages, Dips at ad timestamps are expected and treated as a normal part of the retention curve. YouTube does not specify that these specific Dips negatively impact ranking in isolation from the overall retention signal. What the system measures is aggregate retention performance, not individual point drops.
Should I edit my video to remove sections with Dips?
The official pages do not recommend editing published videos to remove Dip sections as a retention strategy. The retention data from existing views is historical and editing the video doesn't change past retention. Dip data is most useful for informing future videos — identifying patterns of where viewers consistently lose interest across multiple videos reveals structural or pacing issues to address in upcoming content.
Official sources
- YouTube Help — Key moments for audience retention: four moment types and how to read the report
- YouTube Help — Content tab analytics tips: typical retention and reading the retention curve
- YouTube Help — Performance and discovery FAQ: using retention to understand what your viewers want
- YouTube Help — Search and discovery tips: watch time and percentage viewed as ranking signals