A/B testing for advertising is a time-honored technique to check to see which kind of ad is performing well and which one isn’t. Prior to the internet, advertisers would run similar ads in different areas of the same geographic location and use a code for customers to use. That code was then tracked back to each ad, and the ad that had the better results was the one that got used throughout the region. The same principles apply to advertising on the internet and using pay-per-click (PPC), but different landing pages are used to track effectiveness. Following are some tips to help you test your ads and their effectiveness.
What is A/B Testing on the Internet?
Image via Flickr by Images_of_Money
The simplest explanation of A/B testing in the internet era is using two different landing pages for each PPC campaign. One page has a standard version of the layout, while the other page has different features brought forward for the user to view. For example, let’s say landing page A has a prominent picture of the main product, while landing page B doesn’t push the main product right into the view of the customer and instead puts it off to the side. That’s the setup for the PPC campaign.
The PPC campaign comes into play through the use of two different ad campaigns. One ad campaign links to landing page A while another ad campaign links to landing page B. Both ads use slightly different language and/or keywords to differentiate them from each other. The person doing the search might get served up both versions of the ad within the search or will see only one. Either scenario still gives results, as the person searching might not click on one ad but will click on the other because it’s more compelling.
What You Should Test
You can test anything you want. But don’t go crazy and test a bunch of products at the same time. You’ll lose focus, and you won’t get reasonable metrics to show you what’s effective and what’s not. This is a situation where the slower you go, the better your results. And you’re not racing against anyone but yourself, so take your time and do one product instead of many.
You can use your most popular product, or you can try one that’s not doing as well and figure out how to refine advertising on that product to get sales to improve. There’s no right or wrong way to do A/B testing. What you want is information on a consumer’s habits when he or she searches for an item and gets your product or website as a paid ad. Is the ad that you have up a compelling one? Are the keywords right? You won’t know until you get the different landing pages designed and the PPC campaign up and running.
The benefit of focusing on a single product at any given time is that it returns information you can use for a similar product. That information might have results that demonstrate that the ad is broad enough to cover a similar product with very little alteration to the language in the ad. You wind up saving money while increasing the sell-through rate of another product.
How You Should Test
There’s no right or wrong way to do an A/B test. What you need to do is to identify what it is you want to test. Is it product placement on the landing page? Or do you feel there’s an element that’s not getting enough click-through and you want to find out why? This is where you need to do a bit of web design and alteration of elements to find out what’s going to work.
As previously mentioned, you need to first identify what you want to test. Once you’ve gotten that worked out, you need to create a landing page that pushes that element or item in a way that makes it obvious to the average user. How you do that depends on the material design of your page and whether the highlighted item makes it harder or easier for users to navigate. That goes under one campaign, while you leave your existing page under the other campaign. You’ll know what the alteration is, and you can monitor where people are clicking as they land on the page that’s related to either campaign.
Getting People to Click Through
Getting people to click through is another area in which you can test your website for effective keyword usage. When you do a PPC campaign, you’re using keywords to match up to people’s searches. The PPC campaign drops your ad into that search to encourage the seeker to click through and go to your page. If you’ve set up your website properly, you’ve got those same keywords spread throughout your site to encourage a better landing spot on the first page of search engines. This is your opportunity to test how well those keywords are working so you will know whether you need to change them up.
What you don’t want to do is to make the ads the same for either campaign. You need to differentiate the language enough so seekers won’t tune out the ad if they feel like they keep seeing it repeatedly. This is also an opportunity to try new keywords and reconfigure sentence structure to determine how people are reacting.
Never Stop Testing
You can never rest on your success even though you might want to. What you might be able to do is to run the campaign for longer with minimal revisions because there’s no reason to make a major change. But staying fresh and relevant is the key to keeping people coming back for your product, along with bringing in new customers. To that end, keep up with the testing and tweaking to figure out what works, and be willing to make a change when something stops being effective. Remember: change is a constant and little stays the same for very long. The more you can flow with that change and ensure that your online store runs smoothly, the better your results.