Having a relatively high click through rate can have a large impact on performance in your PPC campaigns. Just ask Larry Kim who recently did a presentation on this same topic at SMX Advanced this year. Click through rate is the biggest factor in how AdWords calculates quality score as you can see in the pie chart below referenced by Hal Valarian Chief Economist at Google -
Click through rate (CTR) is 60% of your quality score. It’s critical that your ad stands out from the competition. As a benchmark I always try to have a minimum of a 1.0% average click through rate in my search campaigns. The most effective way to improve CTR is to have a solid A/B testing process and plan in place. What makes up a good process and plan?
Isolate your testing elements and KPI’s
Statistically significant data
Rinse and repeat
Before diving into testing ensure you have a system in place to measure your results. I’m a fan of having some type of shared area to document the starting date and hypothesis for your test. I’ve used Basecamp or even Google Docs. I like these options because they are shared and can be easily changed or updated over the course of your test. I also love using labels in AdWords to label the date the ad was launched. This is a quick and easy way to see the date that your ad test was launched. As part of your framework you should develop a hypothesis. Start with a question – “Why aren’t more people clicking on my ad?”. The answer to this question could be “maybe because I don’t have a free trial”. In this case your hypothesis would be: “Adding a free trial offer in my CTA will increase my ad’s CTR”. Using a hypothesis also makes it easier to communicate a test to colleagues and clients, since most are familiar with this terminology from science.
Isolating Testing Elements and KPI’s
What elements are you testing? Here’s an image of some different elements you could test:
What does success look like in your test? Improved CTR? Conversion rate? Revenue per impression? Or a combination? Make sure everyone involved understands and agrees on the key performance indicators for the test.
Statistically Significant Data
I find this is a VERY important step. Ensure your decision to pause a low performing ad is backed by statistically significant data. There are many tools to do this, I prefer this one by Cardinal Path. I’ve also started using an AdWords script from Free AdWords Scripts to automate ad testing and am really liking it. Sometimes you start a test with good intentions but don’t have enough data to make a decision within a reasonable amount of time. For practical reasons I prefer to end a test after 30 days even if I don’t have enough data for a conclusion. In most cases it’s not productive to have tests longer than this. Impression, click, and conversion volume can be a challenge when implementing tests in PPC with all the different ad groups. To counterbalance the volume issue I like to use an ad copy test template and then slightly personalize the ad across ad groups. Brad Geddes did a great blog post on this topic.
Rinse and Repeat
Testing should be an ongoing part of optimization for your paid search campaigns. I try to get new ads tests going at a minimum of every 30 days but usually volume dictates the testing cycle. As I mentioned earlier just remember to have a shared place to document your learnings from testing. In doing this you’re able to build on past successes and avoid repeating failures from the past. We’ve included a template with instructions that we use at our company to bulk edit and share new ads with clients.