Thursday, August 13, 2015

PPC Ads Not Performing? Reduce Cognitive Load

data-information-overload-ss-1920

If you’re a seasoned online marketer, you’ve probably been exposed to user experience gurus, conversion rate improvement experts and their ilk. Even without knowing the high-level principles, many designers and performance marketers have developed a bias in favor of simplicity. Concepts such as the use of white space and decluttering a page have been bandied around for years in the normal course of creating better experiences for users.

Lurking behind much of the advice are higher-order principles of human-computer interaction (HCI) derived from various overlapping research fields: psychology, cognitive science, ergonomics, ethnomethodology and the like. Top UX experts don’t really feel they “know” anything unless they’ve confirmed certain design and navigation hunches with one of the higher-order principles that may have been tested in other contexts.

For those following the “cognitivist” line of reasoning, we should attempt to link our testing to the existing body of research on how the human brain actually processes information and makes decisions. (Some like to call this neuroscience.)

The Challenges Of Cognitive Load

One of those principles is called cognitive load, which user experience guru Jakob Nielsen defines as the “amount of mental processing power needed to use your site.”

Our intuition tells us that smart people should able to process tons of information and that focused people aren’t easily derailed. That intuition would be wrong. The more distractions we’re forced to wade through, the less likely we’ll be left with sufficient resolve to push through any task, no matter how simple it seems.

It’s surprising how little this concept has spilled over into the creation of the first point of contact many of our prospects have: that tiny little text ad we write to prompt them to take action in the context of their search query.

Pay-per-click (PPC) advertisers have the luxury of trying a lot of ad creative versions and turning them off when they don’t work. But all this spaghetti being thrown at the wall may create little more than a messy house. While unfocused experimentation can turn up unexpected discoveries, it’s not a bad idea to reduce the number of bad or mediocre ideas you’re trying out. Endless testing is sort of the mantra of our industry, but we also need smarter and better testing.

Cleaning Up Your Copy Testing

Assuming our goal is a higher conversion rate, smart advertisers have for years relied on a small number of key principles to guide their ad copy ideation:

  • Particularly relevant advertising. The ads sit at the very beginning — at the trailhead, if you will — of the user journey. Copywriters should be mindful of the well-known UX principle of information scent — which, in this case, refers to users’ ability to predict what they will find when they click from a search ad to your website. Syncing the user query with the ad the user sees — and in turn, to the theme of the landing page — should be a no-brainer. Failure to do so means you’ll experience nastily high bounce rates — or, as Kaushik describes the scenario: “I came, I puked, I left.”
  • Filtering to weed out inappropriate intent. Sometimes, you have to lower your click-through rate (CTR) to make more money. Whether it’s a call to action that refers explicitly to buying or downloading or a line of description that evokes a higher-end or business-oriented purchase, it’s not just the case that winning ad copy does a good job of encouraging; it also does a good job of discouraging the wrong prospects.

However, practitioners of both approaches have forgotten one more thing: Too many ideas (or even characters) crammed into that little ad space could dampen performance by failing to achieve optimal clarity and “cleanliness” in the pursuit of the above two principles. Just as may be the case on a well-designed form or landing page, a little white space could lead to better results.

You’re probably wondering how an ad space so small could possibly produce an ad that is “too long.” Many PPC practitioners spend their time grappling with those 35-per-line character limits. (And we know how that turns out, don’t we? Half the time, the weird abbreviations and tortured word choices perform much worse than plain, simple language.)

Under the banner of “reduce cognitive load,” the UX experts tell us to avoid visual clutter, and they have science to back that up, so why don’t we mix that principle into some of our ads? Here’s an example.

fungus ad

As you can see, the final ad just looks much more cluttered. I find it harder to read. In fact, I’m vaguely repulsed by it, even though my conscious mind might not say so.

But it isn’t just visual clutter. More ideas are being crammed into a small space. The theory says your mind won’t focus on the important ones, or your brain’s executive function will be sufficiently weakened by the experience that your ROI from this cluttered style of ad will turn out to be, let’s say, 15% worse.

The first two ads show us there is a lot of room for experimentation within this general theme of clutter reduction. Both, for example, could be reduced even further if you were fanatical about it.

Note also the elimination of the “helpful” additional “keywords” in the display URL (as shown in the third ad). For years, advertisers have felt this might have a benefit (it might), but have you also considered that it makes your brand name virtually impossible to see or recall without really slowing down and looking hard at it (something most users refuse to do)?

Since CTRs matter, too, a cleaner ad might lead to both higher CTRs (and thus higher Quality Scores and higher ad positions, possibly leading to increased eligibility in the auction) and better conversion rates. That “double win” is the Holy Grail of PPC ad testing.

Build On Existing Mental Models

In addition to eliminating visual clutter and reducing superfluous tasks, the UX experts tell us to build on existing mental models. That principle applies more to your business model, your website’s navigation and so on, but it’s worth considering for your word choices in ads, too. In particular, the principle of building on existing mental models explains why we will usually fail with unusual (or nonexistent) calls to action.

People know what it means to buy, shop or request a free demo. They get confused if you try to sell a service when they think they’re buying a product or if you direct them to a confusing process when they’re expecting a simple, transparent transaction. Be clear.

Failure to follow that principle usually results in severe ROI deprivation and loss of volume. Remember, Google has sophisticated models for incorporating “ad relevance” and “landing page and website quality” into its Quality Score algorithm. Poor user experiences stemming from violated mental models may not be against the law, but they’re likely to be punished by the Quality Score system.

As I ponder this imaginary company with its imaginary vegan deliciousness, I can’t help but think to myself: “I’ve already forgotten the 25% discount. I can see for myself how many varieties are on the page. But what sticks in my head is those robots they have vacuum-sealing those packs! Now that is cool! I am never going to forget Plant… what the heck are they called again? Plantsomething?”

Ah yes, the quest for simplicity never ends. If only the Plantmeisters had been nimble enough to invest in a name like Plants.com or Vegan.com.

As usual in these matters, the last words go to web usability consultant Steve Krug, whose principle is quite simply: “Don’t make me think!”

The post PPC Ads Not Performing? Reduce Cognitive Load appeared first on Search Engine Land.

No comments:

Post a Comment