Sex Up Your Thumbnail: Increase YouTube Views

A loyal reader shared this forum post where YTPaul explained the YouTube changes that have sent many amateur creators into a nose dive (see my Gremlins article).

I quote below, and pay special attention to the bold. Seems those sexy thumbnails are as important as ever!

A thumbnail bait: so retro they work again

Hi all – thanks for your patience as we have been conducting a thorough investigation into those partners that were adversely effected and observed a significant drop in views coming from Related/Featured Videos in early August.

Near the beginning of August, we made a material change to the way in which we select and rank suggested videos that appear on YouTube watch pages.  In fact, it was the most significant change we’ve made to this aspect of our site in some time.  Like many of the changes we experiment with and launch, the aim of this update is to enhance the user experience by displaying the most relevant, engaging content from our vast, rapidly growing catalog.  Simply put, this change is designed to primarily increase the rankings of videos with high user engagement and reduce the rankings of those with lower user engagement. We measure user engagement through a number of factors, primarily CTR, but also include other metrics such as time spent watching video and likelihood a user will watch the video.
The collective results thus far have been overwhelmingly positive.  Users are watching more videos and our suggestions are driving more traffic than ever before.  Despite these positive indicators, though, we realize this change has had a negative impact on some of our partners and uploaders.  With any significant change to our suggestion rankings, it’s simply impossible for each video’s ranking to improve on every watch page — some will go up, while others go down.  After careful investigation, we believe those videos whose rankings decreased on certain watch pages as a result of this change were rightfully demoted in favor of higher-engagement content.
While we realize that our suggestions on the site drive large portions of many of our partners’ and uploaders’ traffic, this is an ever-changing aspect of our site.  To enable more consistent, sustainable traffic to your videos, we are continuing to work on tools to help diversify your traffic sources at YouTube, like our Creator Playbook or the Promoted Videos platform. Ultimately, we believe that continuously working to improve the user experience to drive more traffic through our site is in everyone’s best interests.  We will keep you posted as we roll out new updates and changes to our suggestions and video-discovery models.
Boobs get clicks. Clicks get views. Views get money.

13 thoughts on “Sex Up Your Thumbnail: Increase YouTube Views”

  1. Is there an established link between people drawn to images of cleavage and decent CTR rates?

    Or are you relying on the masses for 1000lurkers’ “idiot clicks”?

  2. What exactly do disloyal readers do? And what happens to them? Sent to the tower? Tortured? Beheaded?

    You managed to completely miss the point YTPaul was making. He implied that videos with gamed thumbnails with resulting poor interaction rates were being demoted because of those poor interaction rates, which the correct thing to do.

    You are so desperate to uncover something that YouTube did wrong that you twisted what you read and somehow came up with him saying that gamed thumbnails were being promoted. The three year old screen cap doesn’t help much either, especially since sxephil did provide links to porn in his video descriptions (a porn promotion posing as a porn promotion isn’t gaming, its TOU breaking, completely different).

    What YTPaul is implying is that you gamed your videos. You shouldn’t get pissed about that. You spent years saying the same thing.

    I know, I know, to the tower,

    but before I go, I would just like to say, I told you so

    Google this and check the date;

    “It is possible to change the YouTube Algorithm to reward honest behavior instead of rewarding dishonest behavior and I believe that if this were done then genuine content creators would benefit.”

  3. If you feel like joining us in 2011, you should have noticed that;

    YouTube just signed up Disney for $15 million dollars worth of original content

    Khanacademy’s $5 miilion grant wasn’t a grant, it was VC money

    Use (abuse?) of all YouTube’s promotional slots is a stated part of the deal with the new TV channels. Unless you are being permanently featured in those slots (like TheFineBros) this promotion is practically worthless now (thanks to the abuse) and YouTube has warned its new TV Partners of that.

    Google+ now has “pages”, so all the performers who added fans to their personal google account are going to have to add them to their “page” as well and update both just in case. In case of what? I don’t know. Its not like anyone actually reads Google+

  4. What should matter more than CTR (driven obviously by thumbnail and title) is engagement rate (duration of view, willingness to comment/rate). Seems like the former matters the most… that’s why my stupid Superbowl video took off despite its poor engagement rate. Until the engagement rate > CTR in influence, then faux thumbnails will always be a problem.

  5. From what I understood, either way, is going to be a battle. I can see a flood of misleading thumbnails, or a number of users doing random giveaways trying to stimulate any kind of response begging for a call to action.

    I guess we’ll have to get clever enough to make people interact rather than just watch (like if that was easy already).

  6. Looks like YTPaul doesn’t know the difference between “effected” and “affected” either.

    I wish YouTube would remove heavily disliked click bait videos from the related videos sections, or at least show the like to dislike ratio along with the view and length.

  7. Can someone answer me this question concerning YT thumbnails: is it something still allowed only for “partners” or is it open for everyone??

Comments are closed.