The PPC Guidebook: How To Run a Split Landing Page Test
The world of pay per click marketing can be a confusing one at times, and you can end up wasting a lot of your ad budget if you don’t know how to use all the features that Google has available for you. Fortunately, we’ve got you covered. Welcome to PPC Guidebook, a Genius eCommerce series on how to run more efficient and more profitable PPC marketing campaigns. For today’s entry into the PPC Guidebook, we’re going to be talking about how to run a split landing page test.
If your PPC ads are placed correctly they’ll capture traffic, but a high click-through rate won’t necessarily turn into ROI if your site doesn’t convert that traffic. All kinds of elements go into that conversion rate, your site’s design, the ease of your checkout process, the quality of your products themselves, etc. The particular element that we’re going to focus on here is the landing page. The landing page is where your potential customers end up as soon as they click on your ad. If the landing page you’re using isn’t relevant to the customers you’re capturing with your PPC ad, you’re going to see a high bounce rate and a low ROI for your campaign.
It may seem simple to pick a relevant landing page for each PPC ad, but Google Adwords provides you a way to be sure you’re using the page that’s best suited to your ad. When you know how to run a split landing page test you can set two possible pages up against one another, and see which one converts better. The more tests you run, the more you’ll be able to hone your campaign, and the better your ads will perform.
How to Set It Up:
The first thing to do is select the campaign you want to test in, and then create a draft campaign which is a duplicate of that campaign. Let’s say that you want to send half of your visitors to my homepage, and half of your visitors to a category page, to see which performs better.
First, you create your draft campaign and name it. Once in your draft campaign, I navigate to your ads.
You select the ads you want to include, then click edit and change ads. You change the final URL to the category page you’re testing against my homepage. You confirm this by clicking Apply.
Up in the top right, you click Apply and choose Run As Experiment. From the Create an Experiment settings you can choose both how long you want the experiment to run for, and how much traffic you want to send to each campaign involved in the experiment. A 50-50 split is typical for an A/B test.
How To Read Your Results:
After your experiment has had time to run, you can take a look at your results. You’ll compare the outcome of your Control campaign (the one you already had running which leads to your homepage) and your Experiment campaign (the draft campaign which leads to your category page). You’ll be able to see which campaign led to more impressions, clicks, and conversions. To test the effectiveness of a landing page you’ll want to compare how many people clicked on the site and how many of them converted in either case. You’ll be able to recognize big changes in outcomes pretty easily, but if the variables are close, you may want to use a statistical significance calculator. Statistical significance is a calculation which is done in order to determine if the results of an experiment are likely to be due to a chance occurrence, rather than the factor of interest that you’re testing. These calculators are available for free online. You can even make your own with formulas in an Excel spreadsheet.
I’ve run the click and conversions in the example above through a statistical significance calculator. The conversion rate is slightly higher in the Experiment group, but not enough to be considered statistically significant. That means we can’t rule out chance as a factor in these results and need to collect more data in order to be certain of what landing page works better.
That’s a quick entry into the PPC Guidebook on how to run a split landing page test. Keep an eye on this series for more guides on running a successful pay per click marketing campaign. Happy bidding!
