YouTube ‘Suggested Videos’ Hint at Good and Bad Titles

YouTube also recommends videos on its homepage.

YouTube publishers who look at the “suggested videos” report in their analytics may find some useful insights about their titles.

On the main analytics page, scroll down to the list of the videos and click on the video title. The link goes to a metrics page for that one video.

Scroll down to the bottom and look at Traffic Sources in the lower left. It will show percentages of traffic under various categories such as:

• Suggested Videos: Views from suggestions appearing alongside or after other videos.
• YouTube Search – Views from keyword searches using the box at the top of the page.
• External – Traffic from websites and apps that embed the video.
• End screens – Traffic from the creator-made end screen of another video or video ad.
• Other – Views from browse features, playlists, direct visits or other features.

The suggested video source is the largest source of traffic for some channels, especially ones in the YouTube revenue-sharing program. (Channels not in the program may have search as the top source.) So it’s an most important one to get right.

Comparing the percentages of each of the above categories between one video and another can reveal some potential ways of improving their views.

How YouTube Suggests Videos

The suggestions appear in the autoplay column on the right as well as at the end of the video if autoplay is turned off.

The video list on the right displays both the thumbnail image of the video and the title. The list at the end of a video displays only the thumbnail unless the viewer waves a cursor over the image, in which case the title appears.

Obviously, a suggested video that appears at the top of the list is more likely to get a click than a video at the bottom.

Relying only on thumbnails to attract clicks has a problem. Various reports indicate that 20 to 40 percent or even more of online users employ an ad block program. It will not only block ads. It also can block the YouTube thumbnails, which leaves only the title to entice the viewer.

Traffic from suggested videos breaks down in two ways much like other content — impressions and click throughs.

The suggested video shows a title and thumbnail. That’s the impression. Some viewers will click and others won’t. That’s the click rate. Publishers of course want to increase both numbers.

Comparisons Reveal Different Results

Video A has 70 percent of its views from Suggested Videos while video B has only 8 percent.

Both videos have been promoted the same. Both videos are part of a series that have content on a similar topic. But the two videos have very different title structures.

Why does video A has so many more views than video B? It’s possible that video A appears much more often in the suggested list than video B. It therefore produces a disproportionate number of views from that traffic source.

It’s also possible that videos A and B have a similar percentage of impressions but that video A has a much higher click-through rate and appears higher in the suggested video list.

The lower click-through rate for video B leads to two obvious questions. One, is the thumbnail not as compelling as the one for video A? Two, is the title not as compelling? In this case, the thumbnails had a similar approach, but the titles were quite a bit different.

If video B has a lower click rate, YouTube may respond by showing it less often or at a lower position on the list of suggested videos. That would send the total impressions, clicks and views into a downward spiral.

After all, why would YouTube suggest the video if people don’t click on the link? Google makes its money from video ads that result from the click.

If a weak title is the cause, then a possible solution is obvious: try a new title.

New Video Title, Better Results

Changing the title for video B (and another) led to promising results. Video B saw a nearly immediate jump in suggested video views from 8 percent to 20 percent. It also had a nice bonus: YouTube search traffic as a source jumped from 29 percent to 34 percent.

In reality, publishers should analyze results over a longer period of time, such as a week or month, and then compare each time period for a trend.

Regardless of the long-term results, the basic point remains. Compare traffic sources between similar videos. Test titles to see if poor performers get a bump.