3 Key Display Advertising Themes For Search Marketers

Digital marketers’ jobs are not getting any easier. Whereas search marketing complexity arises from the sheer volume of keywords and campaigns that brands have to manage and optimize for, display marketing complexity is born from the dizzying array of players in the display ecosystem. It’s hard to keep track of the seemingly endless number of […]

Chat with SearchBot

complexity

Digital marketers’ jobs are not getting any easier.

Whereas search marketing complexity arises from the sheer volume of keywords and campaigns that brands have to manage and optimize for, display marketing complexity is born from the dizzying array of players in the display ecosystem.

It’s hard to keep track of the seemingly endless number of publishers, networks, exchanges and ad tech vendors, each of which plays a specialized and sometimes important role.

And, with the increasing “search-ification” of programmatic display (and social) ad buying, we’ve seen that more search marketers are finding themselves taking on additional cross-channel responsibilities or working more closely with their social and display counterparts.

So, if you’re a search marketer expanding your horizons to include the world of display, what are some of the things that require a shift in mentality?

Giving Conversion Credit To Uncertainty

One of the philosophical challenges search marketers face when managing display advertising responsibilities is that there is an inherent impreciseness that will always exist in measuring display advertising’s impact.

Whereas search advertisers have always enjoyed the position of standing at the end of the fire hose, turning on the spigot when someone requests water and keeping it off when someone doesn’t, this dynamic doesn’t apply to display.

Nobody clicks on display ads. And when people do click on your display ad, chances are they’re not the people you want to target, or they clicked on it by mistake.

But, just because nobody clicks on your display ad doesn’t mean it isn’t making an impact. Thus, the view-through conversion was born.

Why View-Through Conversions Matter

The logic behind view-through conversions is easy to get your head around. If someone sees your ad, that ad should get some credit for the conversion, even if the user didn’t convert on that specific ad impression.

However, view-throughs can still be difficult for search marketers to fully trust. An ad may receive view-through conversion credit, but how do you really know that the ad deserves that credit? How do you know that it had any impact on the user’s eventual conversion? How do you know that the user has even seen the ad?

And, if you agree that a particular ad deserves some credit, what’s the appropriate conversion window for when that credit should be valid? If your customer sees your ad, but makes a purchase two months later, should that ad still receive credit?

Setting Up Lift Tests To Determine View-Through Conversion Validity

The easiest way to try to answer the questions above is to set up lift tests.

The most basic way to set a lift test up is to run an A/B campaign where you run your normal ads in one campaign, and a PSA ad in another campaign.

Once you run the campaigns for a couple weeks, you can then take the difference in view-through conversions between the two campaigns to determine what percentage of your view-throughs are legitimately attributable.

Determining Conversion Windows

windows

The validity of a view-through conversion depends on the conversion window you set. If your window is too long, then you’ll give credit where it’s probably not due; if you set the window too short, you’ll underestimate the potential lift. So what is the right window?

Unfortunately, the answer is: it depends on a lot of different factors.

It depends on what your goal is — for example, leads may have shorter post-view conversion windows than purchases. It depends on your business — for example, B2B solutions companies are likely to have much longer windows than retail fashion companies. It depends on the campaign — for example, a campaign featuring a limited time offer probably should have a comparatively shorter conversion window.

Your best bet is to review your prior conversion data to determine the window where most of your customers go from first impression to conversion.

Is Viewability An Answer For Whether A User Has Seen Your Ad?

As a search marketer, the idea of viewability is a bit strange. “You mean to tell me that some portion of your ad spend doesn’t even have the opportunity to be seen… and you’d still pay for it?” It sounds crazy, but it’s the norm.

What Is Viewability?

Viewability tries to answer the question: did your ad have an opportunity to be seen? For example, if an ad lies below the fold, and the user doesn’t scroll to put it in view, then the impression would be not viewable. However, its scope is limited. It doesn’t reflect any type of user of engagement. You can think of it as a more qualified impression.

The Media Rating Council’s definition of viewability for display ads is whether half the ad was in-focus for at least one continuous second. For video ads, it’s whether half the ad was in-focus, and played for at least two continuous seconds.

So, How Essential Is Viewability Tracking, Actually?

Brands still seem to be treating viewability tracking as more of a luxury than a necessity, to some degree. A recent AdExchanger report found that only 30% of the largest 100 online advertisers are tracking viewability metrics on at least half of their ad impressions. To be fair, that’s twice as many advertisers tracking viewability compared to a year ago — so while not yet a ubiquitous practice, it is gaining in popularity.

Is There A Viewability Sweet Spot?

In an ideal world, you’d prefer to only run ads on publishers and ad exchanges that promise 80-90%+ viewability. That’s how it works on search, right? You’d never pay for a search ad that no one ever sees or clicks on.

Unfortunately, reality is not quite so clean and tidy. There’s a deeper level of faith necessary, not too different from the assumptions you make about the impact of a Super Bowl ad or a Times Square billboard.

Viewability rates can vary wildly based on the publisher, the ad exchange, or which vendor you’re using to track those figures.

  • The percentage of sites that claim viewability rates above 80% is extremely small. If you limit your ad buys to these sites, you’ll restrict your reach significantly and pay higher CPMs. The cost/benefit math of only targeting sites that promise high viewability rates doesn’t usually pan out.
  • You might then think could you hedge your viewability by just buying impressions that are above the fold. Unfortunately, the correlation between viewable impressions and placement on page is weak. Think about your experience reading this article (or any article, for that matter). Was at least 50% of the leaderboard ad at the top of the page in your view for at least 1 second? What about the 300×250 ad in the sidebar next to this text? Which one was actually more viewable?

Again, Is Viewability Tracking Really Necessary?

Viewability tracking is nice to have, but the promise of viewability is greater than the reality right now. The fact is, it’s best not to get too hung up on hitting a specific viewability goal. It’s just drawing an arbitrary line in the sand.

Just like you wouldn’t judge a search campaign’s success based on whether you bid for the #1 ad spot 100% of the time, you shouldn’t determine your display marketing success around a viewability metric.

Focus instead on improving the ROI for specific campaigns by removing the poor performing sites (regardless of viewability) and the sites with blatantly poor viewability metrics, and you’ll eventually get to an optimal cost/performance equilibrium.

You Have Valuable, Exclusive Search Intent Data — Use It!

As a search marketer, you’re in a unique position, because you’re sitting on a treasure trove of proprietary intent data.

Just as you’d run different creative and set different maximum bids for a potential customer who types in [running shoes] vs. [nike lunarspeed lite running shoes], it doesn’t make sense to run the same display campaign to a user who’s exhibited past behavior that would lead you to believe that they are further along the purchase funnel, or a user who represents a potentially higher value customer.

This is why it’s important to work with a cohesive multi-channel platform that can integrate your search, social, and display data, so that you can take learnings from one channel and then apply them in another channel.

What does this look like, practically speaking? At the very basic level, ask the following questions:

  • Can you pipe in your search activity into your display platform and create audience lists based on the search data? You might treat a cart abandoner differently if you knew they came in on a branded search term vs. a non-branded search term.
  • Does your reporting platform accurately measure conversions across channels and devices?
  • Do you have access to audience-level reporting, giving you insight into how specific segments are performing across channel?

Search marketers are in a better position than anyone to wrangle better performance from their display campaigns, thanks to the quality of the data they already possess.


Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.


About the author

Matt Ackley
Contributor
Matt Ackley is a former developer who was dragged kicking and screaming into the world of marketing back in 2005. Now there, as CMO of Marin Software, Matt enjoys creating unique solutions powered by the intersection of science in creativity found in today’s online marketing platforms.

Get the must-read newsletter for search marketers.