Six Pixels of Separation - The Blog
October 11, 2008 9:04 AM

Video Online Does Not Click

With close to ten billion videos being viewed online every month and the average person watching close to eight hours per month, we're seeing many publishers and advertisers jumping on the online video bandwagon.

There's not much going on in terms new advertising models for online video. For the most part, it's a game of pre-roll and post-roll. There are some companies attempting to do lots more. Just this week, YouTube launched their click-to-buy initiative where contextual ads are running underneath the videos with "buy" buttons for music and games (see here - Associated Press - YouTube Flips Switch On New Sales Channel). It sounds like YouTube is turning everyone's video into one long-form succulent infomercial.

How well does advertising in online videos perform? 

Web research company, iPerceptions, leveraged user-generated feedback from 14,000 visitors to some of the leading media sites to gather insights into which types of online advertising gets people to click. The results may surprise you.

"The study found that, despite current buzz around video ads, marketers do not need to spend on fancy interactive ads in order to reach consumers. In fact, consumers are most likely to click on simple text ads (25% of respondents). Display ads follow in popularity, with 20% of respondents likely to click on right banners and 12% likely to click on top banners. A surprising finding of the study is that video ads are not very popular among most consumers; only 11% of consumers said they were likely to click on video ads. And 25 to 34 year-olds show no special affinity for video, being just as likely to click on video ads as text, right and top banners. The only consumers who seem to be engaged by video ads are young people under the age of 25, a group that accounts for nearly one-third of the video-ad viewing audience."

In terms of "what clicks" in online advertising, here is the order (according to this survey):

1. Text ads - 25%.
2. Banner ads (right-hand side) - 20%.
3. Banner ads (top of page) - 12%.
4. Ads in online videos - 11%.

You can read more about their findings here: iPerceptions Study Uncovers Consumers' Real Online Ad Preferences.

The report has an additional fact that is interesting:

"The likelihood that a person will click on an ad goes down as their income rises. On average, 40% of consumers likely to click on any ad make less than $50K a year – and only 15% make over $150K. The income gap is most pronounced with video ads, with 49% of consumers likely to click on video ads making less than $50K a year – and only 13% making over $150K."

Is there a silver lining for online advertising in this economy?

According to Jonathan Levitt, Vice President of Marketing for iPerceptions:

"Our research clearly shows that media sites that offer consumers compelling content and features - encouraging repeat visits - generate much better ad clickthrough rates than less engaging sites. Marketers that want to reach high quality audiences should focus ad placement on sites that deliver the highest customer loyalty and repeat visitor traffic."

If you are building community, providing value and enabling consumers to connect to one another, your advertising and offerings will perform better. Do you agree?

(full disclosure: iPerceptions is a client of Twist Image, I sit on the iPerceptions Advisory Council but Twist Image had nothing to do with this research, and I only found out about it via my Google Reader).

By Mitch Joel


Comments Comments Feed
  • Posted by Jared M. Spool
    Mitch Joel

    Hi Mitch,

    Is there a link to the study? I'd like to see the methodology.

    I think the results are fishy. The big red flag that's jumping out at me is people were asked about their behavior instead of observed.

    Having spent the last 30 years observing people use technology, I can tell you that there's often little match between what people say they've done and what they've actually done. There's even a greater disparity between what they say they'll do in the future and what they will do.

    So, I'm not really sure what this study is saying. From what you report, it's not saying whether people will click on ads. It only says that people *say* they will or won't click on ads.

    Frankly, which do the advertisers care about? What the users say they'll do? Or what they'll actually do?

    Jared

    Reply
  • Posted by Craig Moore
    Mitch Joel

    YouTube may have co-opted everyone's Warhol-ian UCG fame into "one long-form succulent infomercial" but I think we have forgotten that traditional broadcasting is fully and primarily driven by ad supported revenue generation. Tons of great shows have died not because they are engaging and well produced but rather they cannot generate a effective ROI to certain key identified target markets.

    I'd say it's just a transfer of mediums catching up.

    Also in regards to click-throughs, it's all about branding and promotion especially with time-based media. Yes we may be able click on the fly but I think this medium encourages continual streaming which is why product placement works so well and banner clicks work well surrounding the players.

    Craig

    Reply
  • Jared - That was a brain lapse on my part... apologies. I've put a link above to the press release for this news item. From the looks of it, there may not be a study or supplemental information available for download. I have emailed the folks at iPerceptions, and I'm hopeful that they will comment and answer your questions here.

    Craig - I agree with your sentiment on clickthroughs vs. branding, I think the challenge is that online advertising always banked itself on being measurable and action-oriented. As clickthroughs went away, we shifted our stance and now pull the old, "it's a branding channel." I think it needs to be a lot of both branding and clickthroughs for us to really take things to the next level.

    Reply
  • Posted by PIERRE COTE
    Mitch Joel

    great news for those who thinks that advertising goes not with online video...nor either advertising on the web...forget the click and jump into engagement with consumers... merci pour les chiffres mitch.

    Reply
  • Posted by Zelaurent
    Mitch Joel

    Hi Mitch. It’s been a long time since I have commented on your blog.

    This study is interesting, but there is nothing new or surprising about video.

    Video ads are a lot more recent, consequently professionals do not handle it as well as the text and banners ads. The main error is to work with people from media that are used to make TV ads. The catastrophic result of the pre-roll and mid-roll ads is the best example.

    The video interactivity is not exploited yet in online advertising. Let’s wait and see the real potential of video ads before removing it from your marketing strategy.

    Reply
  • Posted by Jared M. Spool
    Mitch Joel

    So, it seems, from the iPerceptions link, that the study was behavioral from site analytics.

    I think the big difference between online video advertising and traditional TV advertising is that the advertiser can now get direct data on when someone has acted on the ad. In traditional TV, measures are based on a ton of assumptions about audience size (which is a guess) and brand strength (which is a guess too).

    I've always thought that all that guessing was hiding some real truths about consumer behavior and the consumer's ability to absorb and act upon ads. I'm thinking we'll learn that people don't think about ads like we imagined they did.

    Reply
  • Posted by Myron Tay
    Mitch Joel


    Does the study also show how the click through rate correlate with bounce rate?

    It seems like a rather good excuse to just label non-working ads as "branding" nowadays.

    Reply
  • Posted by Jonathan Levitt
    Mitch Joel

    Thanks Mitch.
    The details of these results were somewhat counter-intuitive for me. Not the ad data per say, but the demographic data that contextualizes them.
    I'm off to Disney on Ice with the family...my intention was to reply to some of these comments today with respect to methodology (peppered with some of my own opinions on the differences between 'watching' and 'listening'...), but Lightining Mcqueen calls. Keep an eye open for my perspective.

    Reply
  • Posted by Jonathan Levitt
    Mitch Joel

    “The big red flag that's jumping out at me is people were asked about their behavior instead of observed.�

    There is no question based on the ongoing evolution of web analytics and the introduction of complimentary analytical tools, that marketers are moving away from assumption based click path data and looking to contextualize metrics with other equally if not more important data sets (voice of customer etc). The reason is that simply ‘watching’ someone’s behavior or ‘observing’ what they do, is not painting a complete picture. In some cases, in may even be painting a false picture. By ‘watching’, we can see ‘what’ people are doing, but we have no understanding of ‘why’. The objective of this research, was not necessarily to uncover new data on ad preferences (the consumer preferences exist whether we listen to them or not), but rather to provide context into the ‘hearts and minds’ of these consumers, and understand which types of visitors are most inclined to do what we already know based on behavior. Anyone who has ever run a multi faceted online campaign can probably look back at their data and see similar patterns in click through on different ad formats. I know my own data reflects these data quite closely. On that note, I will agree that observing behavior can be quite insightful for usability and user scenario testing, finding glaring usability problems etc, but when it comes to understanding what is in the visitors ‘heart and mind’, and most importantly getting insight on intent, observation will yield very little.

    “Frankly, which do the advertisers care about? What the users say they'll do? Or what they'll actually do?�

    What they actually do, is resulting in very little CTR on most properties. Yahoo! Is offering up a booming average of 0.01 click through and others are not any better. If we want advertising to be more effective online, and ultimately to deliver a greater ROI for marketers, than we need to use the knowledge we gain from what users say they’ll do, test it, and increase what they are actually doing. This speaks to very nature of optimization. Instead of forcing consumers to adapt to what works for the property and brand, why not force the brand to adapt to what works for the consumer? To me, it’s a no brainer. So believe it or not…Advertisers (at least the ones I talk to) care about both, but one is increasingly driving the other.

    Our “self-reported� numbers align perfectly with the results of a study performed by Break Media and Panache earlier this summer. They tested four video ad formats and found a CTR of 10% on pre-roll units without a direct call-to-action. Coverage of the study can be found here: http://adage.com/digitalnext/post?article_id=130388.

    In terms of approach, I don’t want to bore Mitch’s readers with these details, but would be happy to jump on the phone with you (Jared) and walk you through our methodology.

    Reply
Add a Comment

Please complete all the fields below, including the spam filter (to prove you're not a robot).

  1. Fill in your email address to have your Gravatar photo included with your comment.
  2. Please type the word pixels here:
TrackBacks

TrackBack URL: http://www.twistimage.com/movabletype/mt-tb.cgi/925