Last updated: October 2026
Your vlog is live, the views are not moving, and it feels like YouTube never gave it a chance. Before you blame the video, find the stage where it stalled. A video can fail to be shown, fail to be clicked, or fail to hold the people who clicked. Those are three different problems with three different fixes, and the Reach report in YouTube Studio tells you which one you have.
The short version
- Diagnose by stage, not by feeling. Few impressions, weak CTR, and weak watch behaviour are separate faults. Fix only the broken one.
- Impressions measure the offer, how widely YouTube showed the video, not the verdict, whether viewers wanted it. YouTube meters a new video out to a test audience first, then widens or holds back.
- CTR only means something by traffic source. A browse CTR and a search CTR describe different viewers making different choices.
- There is no pass mark. YouTube says half of all channels and videos sit between 2 and 10 percent CTR. That is context, not a target.
- Change one thing, then wait. Judge nothing in the first hours, and give every change time to collect data.
On this page
- Three stages, three different reasons a video stalls
- What an impression actually counts, and what it never counts
- Stage one: the video is barely being shown
- Read the traffic sources before you blame the video
- Stage two: shown often, clicked rarely
- Why the same CTR can be healthy in browse and weak in search
- The 2 to 10 percent band, and why it is not a pass mark
- Stage three: clicks arrive, watch behaviour does not
- One change at a time: the working discipline
- Impressions and CTR questions, answered
- Keep exploring
Three stages, three different reasons a video stalls
Every view passes three gates. YouTube shows the thumbnail, which is an impression. A viewer chooses it, which is the click behind your CTR. Then the viewer stays or leaves, which is watch behaviour. Find which gate your video fails and work only there.
Creators describe all three failures with the same sentence: YouTube is not pushing my video. The sentence hides the fault. A vlog with almost no impressions has a reach problem, and new artwork cannot fix it because almost nobody has seen the artwork. A vlog with plenty of impressions and a weak click-through rate has a promise problem. A vlog that earns clicks and loses viewers in the opening minute has a delivery problem, and that one is not solved in the Reach report at all.
The order matters because each stage can fake the next one. Low impressions make CTR look wild, since a handful of loyal viewers clicking can produce a huge percentage on a tiny base. Work through the stages in order, the way the viewer meets the video.
| What you see | Broken stage | First place to work |
|---|---|---|
| Impressions near zero or far below your usual uploads | Reach: the video is barely being shown | Topic demand, seed audience, how clearly the video signals who it is for |
| Healthy impressions, CTR soft, views low | The click: viewers see it and pass | Title and thumbnail promise, and whether the right viewers are seeing it |
| CTR looks fine, average view duration is short, impressions stall after the first test | Watch behaviour after the click | Retention work, in the guides linked at that stage below |
For the full tour of every report in YouTube Analytics, use our guide to YouTube analytics for vloggers.
What an impression actually counts, and what it never counts
YouTube counts an impression when your thumbnail is shown on YouTube for more than one second with at least half of it visible on screen. Impressions click-through rate is how often viewers watched after seeing one of those counted impressions.
Per YouTube Help, impressions are counted on YouTube Search, the homepage, feeds such as subscriptions and Watch Later, the Up Next recommendations beside the player, and playlists. In the language of the Reach report, those surfaces are where your browse features, suggested videos, and search traffic come from, and playlists feed the playlist traffic source. When a vlogger says a video got pushed, this is the machinery they mean.
Just as useful is what never counts. Help lists external websites and apps, the YouTube mobile website, the YouTube Kids and Music apps, content inside the player such as cards and end screens, email and notifications, background tabs, and thumbnails that flash past too fast or show less than half. Two consequences follow. A notification can send real views without adding a single impression, so views can outnumber impressions. And Shorts work differently: swiping through the Shorts feed produces views, not thumbnail impressions, so a Shorts view count and a long form impression count are not the same kind of number.
CTR has the same narrow base. YouTube calculates it from views that came from counted impressions, not from every view the video owns. That is why dividing total views by total impressions gives a different figure from the one in Studio. Trust the Studio number, and read it in the Reach report for the single video, not only in channel totals where a strong upload can hide a stalled one.
Stage one: the video is barely being shown
A new video is not released to everyone at once. YouTube meters it out to a test audience first and decides whether to offer it more widely. Few impressions usually means a small seed audience, a topic with thin demand, or a video YouTube cannot confidently match to viewers.
Start by confirming the symptom. Compare this video’s impressions with your own recent uploads over the same number of days, not with a large channel and not with your best ever video. A small channel’s normal is a small test. What you are looking for is a video sitting far below your own pattern, or impressions that stopped almost as soon as they started.
The usual causes sit upstream of the thumbnail:
- A small seed audience. Early impressions lean on viewers YouTube already connects with your channel, including subscribers and people who watched similar videos from you. A young channel, or one returning after a long gap, has fewer of those viewers, so the first test is small and slow.
- Thin topic demand. Some vlog topics have a loyal but tiny audience, and some have almost nobody searching or browsing for them. If few viewers ever watch videos like this one, YouTube has few places to offer it. The wider system behind that matching is covered in how the YouTube algorithm works.
- An unclear match. A video that drifts far from what your channel usually publishes, or whose title and subject point at two different audiences, is harder to place. The test audience gets chosen poorly or barely chosen at all.
- A restricted video. YouTube notes that a video judged unsuitable for a broad audience under its Community Guidelines can have its impressions limited. Check the video’s status in Studio before assuming a demand problem.
Notice what is not on that list: the thumbnail. At this stage almost nobody has seen it, so redesigning it treats a symptom you do not have. The stage one fixes are topic choice with real demand behind it, a run of related videos that teaches YouTube who the channel serves, and patience while the test runs. The habits in our YouTube SEO basics for vloggers guide cover the search-led side of that work.
Read the traffic sources before you blame the video
In the video’s Reach report, split impressions and CTR by traffic source. The patterns below are traffic patterns, showing where viewers came from, not a second list of counted surfaces. The source pattern tells you whether YouTube never tested the video with cold viewers, tested it in the wrong places, or tested it well and met a packaging problem.
Open the video in Studio, go to Analytics, then Reach, and look at how viewers found it. A few patterns repeat often enough to learn:
- Almost everything from channel pages and your own subscribers. The video reached the home crowd and nobody else. YouTube has not yet found, or not yet trusted, a wider audience for it. That is a stage one signal, and the fix is topic and audience clarity over the next uploads, not new artwork today.
- Search only, with browse and suggested close to zero. People who go looking find the video, but it is not being offered to people who are browsing. Search can keep a useful vlog alive for years, yet growth past it needs videos that also work as recommendations.
- A suggested spike that dies within a day or two. The video was tested beside other content and the test did not earn a longer run. Look at CTR and watch behaviour for that source before deciding which one ended it: if that source’s CTR held up while watch behaviour fell, watch behaviour ended it; if CTR fell, the promise ended it.
- External traffic doing most of the work. Fine for a launch, but those views largely sit outside the impression system, so they say less about who else should be offered the video.
The same split protects you at stage two, covered next. A CTR blended from every source can look broken when only one source is weak, and healthy while browse, the source that scales, quietly fails. Read the number where the impressions actually came from.
Stage two: shown often, clicked rarely
If impressions are healthy and CTR is weak, YouTube is doing its part and viewers are declining the offer. Work on the promise the title and thumbnail make together, and check whether the video is being shown to viewers it was never made for.
The title and thumbnail are one promise in two parts. The thumbnail has to read at phone size in a fast scroll: one clear subject, a readable idea, nothing that turns to mush at small scale. The title has to name the payoff in plain words a stranger understands without context. When the two repeat each other word for word, you waste half the promise. When they point at different videos, viewers sense the mismatch and pass. The craft side is covered in full in our guide on how to make YouTube thumbnails.
The second cause is quieter: audience mismatch. A vlog about repairing a specific camera model shown to viewers who were watching travel films is being judged by the wrong jury. Their honest disinterest reads as a weak CTR, and the packaging gets blamed for a targeting fault. Clues include a suggested source tied to unrelated videos, and a CTR that is strong in search, where viewers chose the topic themselves, but feeble in browse and suggested, where YouTube chose the audience. The fix is not a louder thumbnail. It is clearer topic signals and a run of videos on one theme so the matching improves.
One warning from YouTube itself: do not chase CTR with clickbait. Help is blunt that a video can win clicks and still lose, because recommendations also weigh whether viewers actually stay. A title that promises what the first minute never delivers buys a click today and fewer impressions tomorrow. To test a new title or thumbnail properly, YouTube’s Test and Compare tool runs versions of one variable against each other and judges them on watch time, which is a controlled test of one change, not several changes at once.
Why the same CTR can be healthy in browse and weak in search
CTR is only comparable between viewers in the same situation. Search viewers arrive with intent and compare near identical options. Browse viewers are cold, scrolling, and one thumb flick from leaving. YouTube warns that CTR varies by content type, audience, and where the impression was shown.
Think about what each audience is doing. In search, a viewer typed a need and scans results that all claim to meet it. Your title carries most of the weight, because it either matches the words in their head or it does not.
In browse features, your thumbnail sits on a home screen among videos from every corner of a viewer’s interests. Nobody asked for your vlog. It has to earn attention against entertainment, news, and far bigger channels. CTR runs differently in that crowd, and YouTube notes the mirror effect too, that a wider offer naturally lowers CTR: when a video earns a lot of impressions, such as a real run on the homepage, a lower CTR is natural, because the offer reaches people further from your core audience. A falling CTR alongside fast growing impressions is often reach expanding, not packaging failing.
The practical rule: compare a video’s CTR with your own videos in the same traffic source at a similar scale of impressions. A niche vlog with a small, loyal audience will often post a CTR a broad entertainment channel never sees, and a wide appeal video will run lower while reaching far more people. Neither number grades the other. Suggested sits between the two cases: the viewer is mid session, so relevance to the video beside yours matters as much as raw appeal.
The 2 to 10 percent band, and why it is not a pass mark
YouTube Help states that half of all channels and videos on YouTube have an impressions CTR between 2 and 10 percent. Read that as a description of the platform’s middle, not as a grade for your video.
That one sentence carries the honest answer. It comes from YouTube, it covers half of everything published, and a band that wide cannot sort good videos from bad ones. A vlog at 3 percent with growing impressions may be healthy. A vlog at 9 percent shown only to two hundred subscribers may simply be preaching to its choir. YouTube adds two cautions in the same FAQ. New videos and channels, and videos under 100 views, can swing wider than the band in either direction. And small movements in CTR are common noise, not a signal to act.
Use the band as a sanity check that your number is not wildly outside what the platform sees, and as a reminder of how much context the number needs. The moment a single percentage gets treated as a pass or fail line, creators start redesigning thumbnails that were fine and ignoring the stage that is actually broken. Your own channel history, split by traffic source, is a better benchmark than any platform wide range.
Stage three: clicks arrive, watch behaviour does not
If CTR is healthy and impressions stall after the first test, the break is past the click. Viewers start the video and do not stay, so the average view duration tells YouTube the offer was better than the delivery. That is a retention problem, and it is diagnosed in a different report.
This article stops here on purpose. Reading a retention curve, spotting where viewers leave, and rebuilding an opening are skills with their own home on this site. The report-by-report walkthrough lives in YouTube analytics for vloggers, and the editing craft that fixes a weak opening lives in hooks and retention editing. Taking your stage three video there is the correct next step, not a detour.
One reach side note belongs here. The promise and the delivery are linked in YouTube’s own guidance: a video with strong clicks but weak average view duration is the clickbait pattern Help warns about, and it earns fewer impressions over time. So when you fix packaging at stage two, fix it toward the truth of the video.
One change at a time: the working discipline
Diagnose the stage, change one variable that belongs to that stage, and give the change time to collect enough impressions to mean something. YouTube advises against judging CTR right after upload and against testing several titles or thumbnails at once, because audience differences then hide the cause.
The discipline is simple and most creators break all three parts of it. They check the Reach report an hour after publishing, when the data is a handful of subscribers and noise. They swap the thumbnail, rewrite the title, and rethink the topic in the same afternoon, so no result can be traced to any single move. YouTube’s FAQ makes the same point: decide only after impressions have accumulated over days, and treat small variations as normal.
A calmer routine for a stalled vlog:
- Wait for a real window. Let the video collect impressions for days, not hours, and compare it with your own uploads at the same age.
- Name the stage in one sentence. Barely shown, shown but not clicked, or clicked but not held. If you cannot name it, you do not have a diagnosis yet.
- Change the one variable that belongs to that stage. Topic and audience clarity for stage one. Title or thumbnail, not both, for stage two. For stage three, the work is retention work in the linked guides, not a Reach report change.
- Record what you changed and when. A notebook line is enough. Without it, next month’s numbers have no story.
- Let the change run to a fair sample, then judge it in the same traffic source you diagnosed. If browse CTR moved and search did not, that is your answer.
None of this makes a single video go viral. It does something more useful: every upload teaches one clean lesson, and the lessons stack. Channels grow on that compounding far more often than on one lucky push, and the wider habits behind it are in our guide on how to get more views on YouTube.
Impressions and CTR questions, answered
Quick answers on impressions, CTR, and videos YouTube seems to ignore.
What is an impression on YouTube?
An impression is counted when your thumbnail is shown on YouTube for more than one second with at least half of it visible on screen. It can happen in search results, on the homepage, in feeds, in playlists, or in the Up Next recommendations beside a video.
What is impressions click-through rate?
It is how often viewers watched your video after seeing a counted impression, shown as a percentage in Studio. YouTube works it out from views that came from counted impressions, so it will not match a hand calculation of total views divided by total impressions.
Why is YouTube not pushing my video at all?
Usually because the first test stayed small: a young or quiet channel offers a small seed audience, the topic may have thin demand, or the video is hard to match to a clear audience. Also check the video’s status in Studio, since a restricted video can have its impressions limited. If impressions are healthy and clicks are the problem, or clicks arrive and viewers do not stay, the broken stage is stage two or stage three, not this one. Compare its impressions with your own recent uploads at the same age first.
Do notifications and shared links count as impressions?
No. YouTube Help lists email and notifications among the places impressions are not counted, and external websites and apps are excluded too. Those routes can still send real views, which is one reason a video can show more views than impressions.
What is a good CTR on YouTube?
There is no pass mark. YouTube Help says half of all channels and videos sit between 2 and 10 percent, and that the figure varies with content type, audience, and where the impression appeared. Compare your video with your own videos in the same traffic source instead.
Why is my CTR high but my impressions still low?
YouTube notes this pattern is common: a video shown to a narrow, loyal audience often posts a high CTR and strong average view duration, while the wider test has not yet followed, as the band section explains. High percentages on a small base describe your core viewers, not a broader verdict.
Should I change my thumbnail if CTR looks low?
Only after three checks: the video has collected impressions over days, not hours, the weakness sits in a traffic source where packaging decides, and the promise is the likely fault rather than audience mismatch. Then change the thumbnail or the title, not both, and give the change time.
Can clickbait raise my CTR and still hurt the video?
Yes. YouTube warns that clickbait titles and thumbnails tend to produce low average view duration, and videos viewers abandon are less likely to be recommended. A high CTR with weak watch behaviour and disappointing impressions is the pattern Help tells you to watch for.
How long should I wait before judging a video?
Long enough for impressions to accumulate over days rather than hours, so the figure reflects a run of days and not a single hour’s swing. YouTube advises against judging CTR immediately after uploading, and notes that new videos and videos under 100 views can swing far wider than usual.
Keep exploring
The reports, the packaging, and the retention craft that sit around this diagnosis.