You’ve scrubbed your keyword list, set geo-targeting parameters, utilized time and day parting, eliminated fraudulent clicks, managed your brand terms and written compelling ad copy. Is there a way to squeeze a few more drops from your search campaign?
Think beyond just bringing traffic to your site and start considering what happens after searchers click. Are you showing visitors what they want to see? Are you interacting with them properly? Are you giving them too many or two few options? Are you even sending them to the right page? So may questions. How can you know? With post-click optimization.
Post-click optimization is the process of analyzing, testing and enhancing landing pages. It should be tackled in phases—Phase 1: Web analytics; Phase 2: Behavioral analysis; Phase 3: Web design; Phase 4: multivariate testing. The process requires a great deal of cooperation between marketing, design and IT departments. And it can be time and labor intensive as well. Yet, post-click optimization has been known to lift conversion rates by as much as 137 percent, an easily justifiable statistic.
Whether you are using a free provider, like Google Analytics, or something more sophisticated, such as ClickTracks or Omniture, you can learn a number of things about your traffic by evaluating your site’s click stream metrics. Web analytics reveals where visitors go, what page they enter and exit on, how long they spend on each page and bounce rate for each page, among other information.
Pretend you are an ecommerce site that sells custom made rubber stamps. You are currently running an extensive PPC campaign with dozens of ad groups, hundreds of keywords in each and several versions of ad copy that follow all best practices. Excellent work. You are also sending everyone to your index page regardless of their search query. Not so commendable.
This is like déjà vu for visitors. They have already conducted a search and believed that you could offer what they wanted better than other advertisers. Why would you make them search again once they reach your site? Instead of putting them through a maddening Groundhog’s Day cycle, you should reward them for choosing your rubber stamps by providing the exact product or feature they searched for, or as close to that as possible. Web analytics is able to match up keyword searches to page views, allowing you to narrow down your landing page to either the most popular or the most relevant one(s).
Based on sales data, you already know that the standard “Received” rubber stamp is the most popular product. By evaluating web analytics, you discover that this product page has the highest enter rate and lots of traffic, yet it also has the highest exit rate. This is the perfect page to concentrate your post-click efforts on.
Now that you have selected a landing page, it’s time to learn how users behave on that page. Heat maps are a great way to do this. They reveal the exact spots where users have clicked—more vivid sections have received a greater number of clicks.
[Insert image: Heat map. Caption: Heat maps reveal where users clicked on a site.]
Back to our rubber stamp example: say the heat map showed a fiery mass over a photograph of the product, the main place that users were clicking. However, the images was static, clicking on it did not lead to another page or generate an enlarged view. Users are easily put off by such inconveniences. Only the most loyal customers will stick around after this nuisance.
Heat maps can also show if customers are clicking on links, headers, and sections of copy or any area of a web page. Heat maps may reveal something you did not realize was confusing, like a bolded word that users click on thinking it is a link, or something that needs modified, such as the static photograph mentioned above.
Now the spotlight trains on the web design department. Based on the analytics and behavioral data provided by the marketing team, your web designers should create a landing page that is user friendly and highly functional. Static images should be linked to enlarged photographs, confusing text must be eliminated or modified, calls-to-action should be clarified and any other changes should be made.
Up to this point you’ve eliminated assumptions and guesswork, and there is no reason to start now. You learned a great deal about your site in the first two phases and made changes based on those finding in Phase 3, but you can’t know if they are actually better unless you test them.
Google offers a free landing page testing tool, Google Website Optimizer (GWO), which allows you to create new versions by placing a small amount of code on your landing and confirmation pages. You can upload numerous versions of different sections, resulting in endless combinations (actually, the maximum number allowed is 10,000, but it’s doubtful that anyone would test that many combinations, not to mention ill-advised; the experiment might run for years without producing definitive results). If your page gets 100 visitors a day, testing two versions of four sections—a total of sixteen combinations including the original—is a healthy experiment that will product significant results and not run indefinitely.
Some page sections that can be modified are headers, font size, length of copy, images, calls-to-action, color schemes, links vs. no links, form location, form length or nearly any other section that you would like to test. Google Website Optimizer mixes and matches the page sections, giving them equal and random play; impressions and conversions are tracked for each page section and each combination.
[Insert image: GWO report. Caption: Combination report from GWO; reveals current conversions, impressions and estimates of future conversion ranges.]
Google Website Optimizer will alert you when significant results have been determined. You can then use the winning combination all the time or develop a follow up experiment to test another page section. The winning combination may not be the one you expected and it could even be one you didn’t like. If the winner strays from your marketing theme or is not the most visually appealing, just remember, your visitors preferred this layout—the customer has spoken and any wise marketer will listen and heed.
Post-click optimization is based on a simple premise that holds up across every industry—give your customers exactly what they want. Learn their preferences and desires by studying how they interact with your site through web analytics and heat maps, and then determine their favorite page layout by conducting multivariate testing. When your visitors are pleased by their experience with your website, they will browse the site longer and are more likely to convert. In turn, the increased sales or leads are sure to brighten your day as well. With rising cost per clicks and increased competition, post-click optimization allows advertisers to squeeze performance out of every click.