Performance Testing for Hosted Ads
Ask any group of online marketers if they measure the performance of the third-party advertisements that appear on their site, or of their own ads, which appear on other sites, and you will likely hear a discussion centered on A/B split tests and clickthrough rates.
There is, however, another major performance metric that has a significant impact on revenues, but is often overlooked: load times.
Take a complex, jQuery-driven website that requires pages to be fully loaded before being displayed, for example — holding up the page’s final rendering until an iframe or other element is ready for use. If the ad’s server is down or operating slowly, then your site will be down or slow as well. Even static sites are susceptible to load-time induced performance issues, which are becoming more common as webmasters increasingly rely upon third-party content and ad network injections to power and profit from their sites.
Likewise, if the ad network serving your ad is not up to par, you may see impressions being recorded (and charged for); but if that ad didn’t load before the visitor left the page it was displayed on, then it didn’t do you any good — but you still had to pay for it...
Perhaps the poor CTR your campaign is experiencing is due to incomplete loading, rather than any problem with the quality of the creative.
While careful coding, including the strategic use of “document.ready” and “onload” commands will help mitigate the harmful effects of slow-loading advertising, techniques such as image swaps (an old-school standby from dial-up days, where a low resolution placeholder was initially loaded, with the actual image replacing it once it had loaded), and staggered content displays can also be used to develop a workaround path.
For example, to combat a slow-loading “live models” iframe ad, this author first loads a page containing a static ad or image into the iframe, using a meta-refresh to display the live cam ad once it has finished loading. This technique doesn’t actually boost the page’s performance, but it does give the viewer something to look at while the page fully loads — providing a perception of seamless content loading.
The IAB offers a series of best practices for boosting advertising load performance (www.iab.net/adload), while an innovative load-testing service known as LoadStorm (www.loadstorm.com), provides comprehensive and tangible cloud-based analytics.
According to LoadStorm, the impact of hosting third party content on a website’s performance can be summed up in two words: “enormous” and “detrimental,” with even a single third-party widget slowing a web page’s load time by more than 250 percent — highlighting the need to understand and optimize these huge performance roadblocks.
LoadStorm’s Scott Price says third-party content has advantages and disadvantages.
“On the one hand, hosting third party content enables sites to add functionality that might otherwise take them months or years to build themselves,” Price wrote. “On the other hand, that additional functionality comes at the cost of performance.”
“The more third party content a site hosts,” Price added, “the greater the risk it runs of sluggish load times, request timeouts, and client-side parsing errors.”
Thorough testing will reveal whether or not the value of a widget is high enough to justify the poor performance trade-off and its well-documented negative impact on sales.