Click Through Rate guide - Exposing YouTube’s BIGGEST lie

Which YouTube Impressions Click Through Rate would you rather have? 4% or 10%?
10%, right?
Because when 10 viewers out of 100 click, results in more views than when 4 people out of a 100 click.

Click Through Rates (or Impression-Click-Through-Rates, commonly abbreviated as CTR) is one of the key metrics that the YouTube algorithm looks at to determine whether it should promote your video or not.
However, Click Through Rates are misleading and comparing them is next to impossible. Because CTR doesn`t tell what you think it does.

A higher Click Through Rate doesn’t always equal more views.

In the example above:

Video 1: 10% CTR: 7K views
Video 2: 4% CTR: 330K views.

What is the meaning of Impressions Click Through Rate (CTR) on YouTube?

Click Through Rate (or CTR) is the percentage of viewers that see your title and thumbnail and click on it.
The CTR is expressed in a percentage.

Click Through Rate is a broad marketing term and not exclusively used in the context of YouTube.
Also for online advertisements and links in marketing emails are Click Through Rates measured.

YouTube uses Click Through Rates combined with Average View Duration, Average Percentage Viewed and returning-viewers as the most important metrics to promote a video in the algorithms like Suggested, Home and Search.

Why is CTR not reliable?
Besides the above mentioned metrics, there are more metrics that YouTube looks at: like/dislike ratio, time on watch page, amount of comments and probably 100 other factors that YouTube won’t reveal.

Herein lies also the problem for using CTR as a reliable metric.
Click Through Rate is only one factor for the algorithm to consider whether it promotes your video or not.

How is YouTube Impressions Click Through Rate calculated?

The formula for calculating Click Through Rate:
CTR % = (clicks / impressions) * 100

CTR = Click Through Rate measured in percentages.
Clicks = how many people click on the title+thumbnail
Impressions = how many people have seen the thumbnail.

If 100 people see your thumbnail and 6 click on it, you have a Click Through Rate of 6%.
If 18,134 people see your thumbnail and 578 click on it, you have a Click Through Rate of 3.2% (=578 / 18134 * 100).

What counts as an impression?

In theory YouTube measures if a thumbnail is shown on a page where the viewer would have been able to see it.
But YouTube doesn’t (or can’t) 100% accurately measure every Impression.

More specifically:

The thumbnail has to be shown for more than 1 second
The thumbnail needs to be visible at least 50%
On YouTube, an impression is counted right away if a viewer clicks on the thumbnail.
Clicking on a link on a website or app does not count as an impression.
Clicks from notifications and end screens are not included
If a thumbnail is shown twice to the same user (even if they`ve watched the video already), it is counted as 2 impressions.

Sources: https://creatoracademy.youtube.com/page/lesson/analytics-impressions and https://support.google.com/youtube/answer/7628154?hl=en

Why Click Through Rate is unreliable
But if YouTube doesn’t count impressions for every traffic source, can you make a reliable comparison between click-through-rates of videos if the traffic sources are not the same?

What is a good Click Through Rate on YouTube?

From my experience with informational content a CTR between 2-4% is quite normal and
4-10% I consider high.
My highest Click Through Rate on a video ever is 14.1% over 28 days.
But regretfully a good Click Through Rate doesn’t equal a lot of views.

YouTube itself says: A CTR between 2% and 10% is considered normal on YouTube:
• 10% is high, meaning a lot of people click on it
• 2% is low, meaning less people click on it
I’ve seen people have CTRs of 18%, but that is not always a good thing, as I will explain later in this article.

General topic videos (i.e. humor or vlogs) tend to have lower Click Through Rates than targeted audiences (i.e. people who like dancing, race car fans or people who like to paint).

I have listed the traffic sources in order of higher CTRs:

channel page – most interested in you
YouTube search – interested in the topic
YouTube homepage – maybe not interested in the video topic

You can find your CTR by traffic source in your YouTube Analytics.

CTRs from your channel page are probably higher than CTRs from YouTube’s homepage and search.
This is because the target audience is more focused.
Someone that looks at your channel page is probably more interested in you (and thus more likely to click on one of your videos) than someone that is seeing one of your videos on the YouTube homepage.
But even on your channel homepage, all your own videos compete with each other.

Why Click Through Rate is unreliable
The Click Through Rate given in YouTube Studio, is an average number across all traffic sources.
But the individual traffic sources tend to have percentages that are good and bad for that traffic source.
On top of that, videos don’t tend to get the same ratio of traffic from the same traffic sources for every video.

That makes comparing Click Through Rates between videos cumbersome when you want to do an apples-to-apples comparison.
You have to take the traffic from the different traffic sources into consideration.

The data doesn’t lie
I’ve collected the views, impressions and click-through-rates of 50 videos on one of my channels.
I want to show that a high click-through-rate does not relate to a high amount of views.

Views	Impressions	click-through rate %
537474	7073079	5.38
342167	5241719	4.34
227665	2355352	7.26
197504	1992871	7.43
149983	2259864	3.97
147955	668250	12.82
133780	1506164	6.18
127861	1278225	5.68
111056	886288	7.74
100999	1032816	5.58
100614	962177	5.97
97856	801052	8.79
97459	1194262	4.71
87782	924529	5.47
86365	1189264	4.48
80492	505522	8.08
79983	307669	8.76
78701	612039	7.15
72794	418170	5.64
72328	738398	6.43
71123	741635	4.96
64135	581191	5.62
61490	444476	9.22
56839	320492	10.76
55249	357884	7.96
52846	395493	7.31
51537	431011	4.05
50410	622274	4.52
49761	556694	6.41
47183	453351	6.41
44757	301637	6.49
43832	487046	4.69
42258	349107	7.8
38138	259241	8.67
37826	402410	6.26
37246	394064	6.15
36946	229291	10.15
36872	264775	9.07
35326	330389	5.18
34719	352466	5.33
33990	296342	8.3
32943	176406	9.01
32779	219639	9.55
32639	127466	14.34
32199	177867	8.21
32156	186305	8.24
31702	127075	8.76
31655	191377	7.97
30562	261499	8.46
28771	163016	11.46

Why you can`t compare Click Through Rates on YouTube

I’ve stated a couple of times that you should be careful with comparing Click Through Rates.
Here are some additional reaons:

Click Through Rates are relative to the competition.
You don`t know the Click Through Rates of the competition.
If the competition has CTRs of 10% and you 9%, you will “lose” in the rankings, while 9% is a good CTR.
Click Through Rates are different for different types of content and topics
Click Through Rates tend to get lower over time (not always).
Look at a couple of your videos and look at the change of a CTR over the time span of a couple of months.
CTRs for new videos are unreliable.
The algorithm is busy testing the video to different audiences and different places, so the CTR is all over the place.
CTRs for small channels and videos with not a lot of views are unreliable.

The data lacks to be statistically significant.
You need to have at least 1000 views on a video to get a somewhat reliable CTR (preferably even 10K views).

The only fair comparison between videos is on a large channel with videos of the same age, same ratio of traffic from traffic sources and on the same topic.
That is really hard! (read: almost impossible).

The best comparison you can do on a video itself (i.e. changing a thumbnail for example), because the audience stays (somewhat) the same.
But a better comparison would be to look at the amount of increased/decreased views over a multiple weeks.
Or even better, look at the changed watch time, because changing a title/thumbnail can change the viewer expectation.

Why a high Click Through Rate is not always better

You would think that a higher number is always better.
The funny thing is CTR says “nothing” about how bad/good your video performs.

It only tells you how bad/good it performs with the audience that the YouTube algorithm suggests your video to.
Over that last part (“the audience that the YouTube algorithm suggest your video to”) you have absolutely no control.
CTR is a relative number, relative to the audience YouTube picks for you.

Videos that have a lot of views, have generally a lower Click Through Rate.
(Read that sentence again!).

This has everything to do with your audience.
If your video performs well (in terms of Average View Duration, Watch Time, CTR, returning viewers, etc.), YouTube will suggest your video to a broader (less focussed) audience.
A less focussed audience is less likely to click on your video, meaning that your CTR will drop.

The process repeats: if that audience is happy, video is suggested to a broader audience, if that audience is happy, broader audience, happy?, broader audience!, etc.
Since the YouTube algorithm will keep suggesting your video to a broader and broader audience until it hits the sweet spot, your CTR will always get a hit when it hits a new target audience.
You can see these “steps” and spikes clearly when you look at the CTR graph your YouTube Analytics.
If that happens, you know that your video is growing.
If that doesn’t happen, your video views will stay consistent or go down.

That means that Click Through Rate fluctuates, because YouTube is constantly testing the video on different audiences.
Click Through Rates are different from one day to the next.
CTR is particularly unreliable for small channels, because 3 views can mean the difference between a super high and a super low Click Through Rate.

Doing a/b-tests on thumbnails and looking at a CTR at a certain day doesn’t give you reliable data.
You should do that over an extended period of time with enough data to make the comparison statistically significant (at least a 1000 views).

Also seasons or events can trigger fluctuations.
Around Christmas there is a renewed interest is Christmas trees, where’s in June not at all.
The same goes for topics like Superbowl or the Olympic games for example.

Successful YouTubers have a lower Click Through Rate than smaller YouTubers because their videos are being suggested to a broader audience.

How to read Click Through Rate then?

How do Impressions, Click Through Rate, Average View Duration (AVD) and views relate?

Impressions	CTR	AVD	Views	Meaning
Low	Low	Low	Low	Bad thumbnail and/or title People don’t like your content
Normal	High	Low	High	Clickbait (views and impr. won’t last long)
High	Low	High	High	Broad audience (i.e. vlogs, comedy etc)
Low	High	High	Low	Specific audience (i.e. music producers)
High	High	High	High	Viral video

Let’s break some of these down:

High CTR + bad AVD
If you have a high CTR and bad Average View Duration:
People that click on your video don’t like the content they see when clicking.

That is probably caused by your title and thumbnail communicating clearly enough what the video is about.
It could also be that the hook in the video intro does not grab peoples attention.
In other words: your video is not what the viewer expects when they clicked the thumbnail and clicked away.

If you did this intentionally, it is called clickbait.
Be very careful with High CTR and bad AVD, because YouTube stops promoting your video entirely, the algorithm thinks you try to trick people in watching your video.

My article on improving Audience Retention will help you out if that is the case.

High CTR + good AVD
If you have a high CTR and good Average View Duration, you could try to change your video title or thumbnail to see if you can make the transition to a broader audience and get more views.
But, be careful, you can also hurt the performance.
If the views drop you should obviously reverse the change.
Always save your old titles and thumbnails in case you need to go back in time.

Never ever change title+thumbnail from a well performing video.
It can only be worse.
It will only hurt you more than you will gain.

Low CTR + good AVD + low views
A low CTR, good Average View Duration and low view count can mean 2 things:

Your topic is not appealing to viewers (too niched down for example)
Your topic is appealing but your videos aren’t getting clicks.

In the last scenario you should definitely change your title and thumbnail.
No harm, no foul, you have nothing to lose.
I’ve had success with this strategy in the past.

How do you increase Click Through Rate on YouTube?

Assuming that you want lower (not higher) click through rates because you want more views, here are some tips:

1. Title and thumbnail work together

Both title and thumbnail are important to get the click.
An amazing thumbnail might perform well, but if viewers can’t make head nor tails of your title, they won’t click (and YouTube won’t promote).
The same applies the other way around: nobody clicks great title with a bad thumbnail.

The title and thumbnail are more important than having an amazing audience retention.
Having a good title and thumbnail can make large channels out of small ones.

2. Make it stick out.

You are up against at least 20 thumbnails at the same time for a viewer.
Make sure it sticks out.
If everyone uses blue in their thumbnail, use red.
Use a photo editor to paste your thumbnail design into a screenshot of the YouTube homepage or YouTube search results to see if it sticks out.

3. Run thumbnail through Google Cloud Vision.

YouTube uses a version of this algorithm to determine contents of the thumbnail want to reduce the spammyness to increase likelihood of success.
You can find Cloud Vision here: https://cloud.google.com/vision/docs/drag-and-drop .

4. Improve audience retention

Although a good Audience Retention is not as important as the thumbnail, a bad audience retention certainly hurts your views.
Read my article on How to improve Audience Retention for more details.

Good storytelling is vastly underestimated technique to keep viewers hooked (also for educational content).
In my article on the storytelling formula I’ll teach you all the little secrets you need to know.

5. Ask feedback

Post multiple variations of your thumbnail on your community tab and ask for feedback.
You can do this also on other social media and your email list.
It helped me to gain valuable insights that viewers with a fresh eye looking at the matter.
As creator you are too deep in to form an objective opinion about your own thumbnail.

6. Look at titles and thumbnails of other videos

Look at other videos on your niche that perform well in terms of views.
Are there commonalities in well performing videos in terms of color, facial expressions, topic??

Also consistent thumbnail elements like style, font, logo, face and colors can help viewers to recognize your thumbnails.
This will improve your video in the recommended videos algorithm (“if the same people click on your videos over and over again, it must be a good video”).

7. Know your target audience

Who do you make the video for? Age, gender, location, family situation, dreams, struggles, financial situation etc.
You can make more educated decisions for title and thumbnail, but also video topic.
If you make the video for Jane, 23 years old, interested in fashion and has a boyfriend, you will make a title and thumbnail that resonates with her (in color choice, word choice, topic choice).
When you are serious about YouTube and want to take it to the next level, take a look at my video training program: Viral Strategy.
The program takes you step-by-step through the process of getting views, subscribers and going viral.

For new creators I included a module that guides you step-by-step through the process of starting, creating and setting up a YouTube channel.