Google Ads A/B Tests are the holy grail of pay-per-click (PPC) advertising for search. Without proper A/B tests, it’s difficult to get a read on what changes are actually working for your business.
Here are a few considerations to layout before launching any Google Ads A/B test:
1. What is the purposes of my test?
Every test you run should have a clear, defined purpose that you are trying to achieve.
For example, you may want to run ad tests to improve click-through-rate (CTR). We run Google Ads A/B tests for ad units constantly as CTR has been found to be a heavily weighted component of Quality Score.
Remember, Quality Score is made up of 3 main components: CTR, Ad Relevance, and Landing Page Experience + the use of ad extensions. Improvements to Quality Score can drastically decrease your costs-per-click, increase your impression share, and improve your profit margin.
Another example includes landing page testing for conversion rate optimization. We often like to look at how category vs product pages perform for a select group of keywords.
Running landing page A/B tests ensures that we are landing website visitors on the right landing page to yield the highest conversions.
Or, you may want to run full Google Ads account structure tests or campaign settings tests to see if Google’s new features actually work the way the say they do. Recently Google Ads launched the eCPC setting for bid management and the Optimize setting for ad rotation.
You may want to look at how your historical settings stack up against Google’s new features before diving into a full campaign change.
Subscribe to Trent’s newsletter to stay up to date on all the latest performance marketing posts.
2. What key performance indicators (KPI’s) am I going to measure?
Clearly defining your KPI’s before a test launches helps to ensure that your Google Ads A/B testing roadmap is on track to deliver valuable learnings from what it is you are testing.
Sometimes tests may fail and that is okay. The key to is to work out the hypothesis in an effective and efficient manner, then record the learnings and move on to the next test. Think wash, rinse, repeat.
We like to keep our tests simple and look at 4 main Google Ads metrics that can tell us a lot about the test we are running: Clicks, Impressions, Cost, and Conversions
These 4 metrics provide you with enough insight to examine what the click-through-rate (CTR), cost-per-acquisition (CPA), and Conversion Rates are between variants.
Here’s an example of what that might look like with mock data:
As you can see, variant B has a higher click-through-rate (CTR) and lower cost-per-acquisition (CPA). In this simplified example, we would most likely go with variant B if it reached statistical significance.
Kissmetrics has a useful statistical significance calculator that you can use to check to see if your A/B test has reached statistical significance or not.
3. How am I going to set a true A/B test up?
The easiest way to set up true Google Ads A/B tests is to use the Drafts and Experiments feature. Drafts and Experiments allows you to split your traffic evenly in order to get a read on which variant performs better for you and your business.
It also serves as an opportunity to double check which variable you are testing. Single variable tests are critical for learning throughout the testing process.
Far too often we run into PPC accounts that are testing multiple variables within a single test. For example, they may be testing different ad headlines and different landing pages at the same time. This makes it difficult to understand which variable actually led to the change in performance.
Was it the ad headline or the landing page?
It is super important to be laser focused on what specific, single variables that your Google Ads A/B testing roadmap is attempting to test to avoid weeks or months with little to no learnings.
To set up Google Ads A/B tests in Drafts and Experiments using the old UI:
1. Click into the campaign you are trying to test
2. In the top right corner, toggle the Drafts dropdown and select “Create new”
3. Create a name for your draft and click submit
This will take you to the Drafts section of Google Ads where you can make the changes that you are trying to test.
4. Once you’ve made your changes, Click the Apply button in the top right corner and Select “Run an Experiment”
5. Enter a name for your experiment, select a start and end date, and select how you want to split the traffic
We recommend running 50/50 splits to give each test cell enough data to see performance changes.
6. Click “Create”
After your test has began running, you’ll be able to monitor performance within the Experiments tab of Google Ads by clicking into your experiment campaign:
To set up Google Ads A/B tests in Drafts and Experiments using the new UI:
1. Click into the campaign you’d like to test
2. Click the Drafts & Experiments tab on the left sidebar and click the (+) icon to create a draft
3. Enter a Draft name and click SAVE
4. Once your draft is created, make the changes you’d like to test to the draft version
5. Navigate to the Campaign Experiments tab
6. Select the draft you just created, enter a name, select a start and end date, and your traffic split
7. Click SAVE and your experiment will be created
Once your experiment has began running, you’ll be able to monitor performance by clicking into the experiment campaign:
4. Monitor and track your tests
Once you’ve established the purpose of your test, defined your KPI’s, and set up a true, single variable A/B test using the Drafts and Experiments feature, it’s time to report and track your testing progress.
Our A/B testing roadmap is simply a running list of A/B tests that we have ran documented in a Google Sheet. This helps us annotate what happened, what was the learning, and what we plan to test next.
Monitoring and tracking you’re Google Ads A/B tests helps you and your team understand why you’ve made changes in the past and why the account is set up the way it is. This can also serve as a historical guide that can be used as first point of reference for future testing hypothesis.
Constant A/B testing is critical to the success of your PPC advertising program. With the proper testing structure, you’ll be able to unravel what works for your business by using hard earned data that you can capitalize on.
Subscribe to Trent’s newsletter to stay up to date on all the latest performance marketing posts.
Continue Reading: WHAT IS PPC ADVERTISING?