It took me a while, but I finally got to one of the key items on my summer reading list: DoubleClick’s report on The Brand Value of Rich Media and Video Ads. And I wasn’t disappointed by what I found. For the first time in years, DoubleClick dug into their server logs and gave us a bit of hard data on what types of rich media are being served. They also worked with Dynamic Logic to conduct real live brand surveys (rather than tracking those overly simplistic ‘engagement’ metrics so many rich media vendors have become enamored of) to study the brand metrics that matter to advertisers: things like brand awareness, brand favorability, message association, and intent to purchase. And what they found was sometimes surprising:
- There’s a lot of rich media being served – but most of it’s not very rich. DoubleClick says that in 2008, less than 40% of the impressions it served were GIFs or JPEGs. So does that mean the other 60% were rich media? Not according to DoubleClick: it says 55% of its 2008 impressions were “simple Flash” ads. Just 6% of the impressions served in 2008 were rich enough to be labeled “rich media.”
- Rich media ads that incorporate video are usually your best choice. This creative format was most effective on nearly every brand metric, including aided and unaided awareness, brand favorability, and purchase intent. The only time not to use rich media with video? Surprisingly, when you’re looking to build message association an old-fashioned GIF or JPEG is by far the most effective format.
- If your Flash ads don’t include video or interaction, don’t bother. When you add up performance across all the brand metrics studied, simple Flash ads provide less brand impact than any other format – even GIFs and JPEGs. It turns out all those advertisers who served simple Flash ads through DoubleClick last year could’ve saved themselves some time and hassle by simply producing animated GIFs.
If you’ve had a look at the report, let me know in the comments below what you found interesting or surprising – and whether you’ll change your creative strategy based on the findings.
Not that simple Nate. DCLK’s report is so vague, I am actually a bit surprised you find any real value in it.
There are so many ways to use video with display – Is the video being used for visual effect only? …to demonstrate something detailed? …to convey branding message? …to drive direct response? …how long is the video? …is audio required? None of this is covered in the report.
Do we really need a DCLK report to tell us that static gifs with the key brand messaging and call to action present immediately and at all times, provides better message association than a Flash banner with :15 seconds of animation that delays the brand messaging and call to action until each and every frame plays out? Most users don’t even stay on the page for :15 seconds or they have scrolled down below the fold while the animation continues to play out of view. For any ad to be effective, the key messaging and call to action should be available throughout the animation. In any given frame, the user should know what the ad is for and what they are suppose to do.
The major challenge is that most agencies do not know how to use display effectively, whether branding or direct response action. “Learn More” is not a call to action. Agencies make big $$$ building websites, not banner ads. The skill set required to design effective banners is very different than websites. It’s the difference between long form and short form. Unfortunately, very few creative agencies understand this, let alone how the formats they build impact metrics and results. Media teams are so focused on audience, social, targeting, and data that they know very little about creative. In most cases they just have not seen enough creative to know what “works”.
DCLK’s sweeping generalization about Video and Flash v. gif is a major problem for the industry. Everyone is looking for the “silver bullet”. The reality is DCLK has done very little to advance creative success over the years and has actually diminished it’s value. Instead it focuses on ad serving, targeting, and data solutions. DCLK has done a terrific job marginalizing the value of the creative design and format as success drivers. This report is another perfect example and proves how little DCLK actually understands about the gazillions of ads they serve every day. Sure they have some decent creative design tools, but they do not help their clients use them effectively and I don’t think they care.
Nate,
video is something that has always been a key component of rich media that influences data points. Think from a creative viewpoint, most people ‘play safe’ with rich media for fearing publisher kick-back, or locked into a mindset of ads=clicks as opposed to ads in any medium are a way of influencing and changing perceptions, i.e. the true definition of engagement.
Wondered if you had taken a look at the research I’ve been involved with on a global level in measuring rich media as a time-based component in order to address the floors inherent within Interaction Rate? There are some exciting new findings re: video and rich media in there…
http://bit.ly/Kp0xL
Let me know what you think.
Lee brings up a great point. There is an overwhelming amount of bad creative out there generating bad results. And I don’t put the “dancing coyboys” and “pigs with 50 states on their backs” into this category because dollar-for-dollar those probably get good results, or the advertisers wouldn’t continue to pay for them. I’m talking about creative that is supposed to bolster brand awareness but doesn’t display the logo until 10 or 15 seconds after the page loads. Or expandables that don’t have a call to action to expand. Or video that doesn’t clearly ask the user to un-mute.
OK great, glad we all got that off our chests. Now do we need DoubleClick to tell us that rich media works in aggregate across a wide variety of impressions? Yes, as long as I keep hearing from Techcrunch, the New York Times, and others that “display doesn’t work” or that “no one has proven that display advertising has an impact on the Internet” I think we do.
As for Dean Donaldson, I’d recommend following him on Twitter and letting him buy you drinks. I’ll leave the comments about his research to others.
Once more we see reports that measure what’s easy to measure and marginalize what’s hard to measure. Computers are calculators at heart, after all.
Everyone with any experience knows that creative and media must go hand-in-hand to ensure success. The keys are execution and context — the right ad in the right place at the right time. Regardless of media type.
I read the Doubleclick report like a parlour game — a topic for conversation but little value for implementation.
From my experience, people deferring to industry standard reports are just not doing their jobs. They’re not experimenting to find out what works for them, for their context, for their business. They’re hiding behind best practices.
James, if you think it’s easy to measure brand impact, then clearly you haven’t had the distinct pleasure of implementing an online branding survey lately. They’re a pain in the ass, and I’m glad whenever someone is willing to put in the time, the effort, and the money to collect an entire body of branding research, as DoubleClick has done here.
I also think you’re being overly harsh when you talk about the value of standards. Of course marketers should do the legwork to find out which formats and creative approaches work best for their products and their customers – just as, in the long run, they should benchmark against their own previous campaigns rather than against industry averages. But every marketer needs to start somewhere. And given what our surveys say about how few online brand marketers measure their success, it appears most are still much closer to the starting blocks than to the finish line.
If a marketer is still deferring to industry benchmarks and best practices after a year of running (and properly measuring) rich media and video ads, then they’re foolish; but if they *don’t* defer to industry benchmarks and best practices when they’re starting out (or when they haven’t been sufficiently measuring those ads themselves, as most currently don’t), then they’re equally foolish.
Good points all around, Nate.
You’re right I’ve never had to measure brand impact, and I don’t think I’d ever want to have to. Seems like juggling Jell-o.
Overly harsh on the value of standards? Maybe. But none of this is new and the numbers of people who make decisions based on, well, what exactly? is astonishing.
I gave a talk a few years ago titled ‘Marketers make decisions based on faith, not reason’ that I hoped would provide a little guidance for iterative, data-driven decision making. But no, not so much.
Sorry if my annoyance with marketeer laziness came our harshly.