Ads are the only part of your account a searcher sees. They set the expectation for what the user will find on the landing page. When we think about the basics of testing, we can test in three ways:

• Ads versus other ads: This tests the message the user sees
• Landing page versus landing page: This tests the landing page that the users see for the same ad message
• Ad & landing page combination versus another ad & landing page combination: This tests the ad and landing page interaction

Let’s take a brief look at how to set these up and the metrics to use.

Ads versus ads

An ad versus ad test is simple. You first choose whether to use multi ad group or single ad group testing (more info on that here).

Next, you create your test and pick winners based on your testing metrics, which are determined by your overall goals.

In general, conversion per impression or revenue per impression metrics are best. With impression-based metrics, we’re looking at opportunity. The ad being shown to a user (impression) is the opportunity, so it’s best to measure from the impression.

Landing page versus landing page

In landing page testing, you have two options:

• Test two entirely different pages. This can be accomplished through simple ad final URL manipulation
• Multi-variant testing. This requires landing page software to accomplish

If you are testing multiple pages, then you can create two ads:

• Ad 1 goes to page 1
• Duplicate the ad and change the Final URL to page 2

This test will send the traffic to two different pages using the same ad copy, and the ad data will show you which combination is best.

With landing page tests, you don’t want to use impression-based metrics as the user doesn’t know the landing page before they see it. Here, you want to use conversion rate or ROAS (as pages may differ in both quality score and average checkout amounts) as your testing metrics.

If you’re testing landing page templates and the base template is being used in many ad groups, you can also use multi-ad group testing. This method shows which template is performing the best across your ad groups.

Landing page & ad combination testing

This is sometimes called customer journey testing. You’re testing how two different pages and two different ads act in combination with each other. Even this testing is simple to setup:

• Create ad 1 and send it to page 1
• Duplicate ad 1 and send it to page 2
• Create ad 2 and send it to page 1
• Duplicate ad 2 and send it to page 2

This test will show how each combination of ad and landing page message interacts with each other.

With this test, you can use ad data to determine your winning ad and landing page combination. As you’re testing the entire customer journey, you should use impression-based metrics, such as conversion per impression or revenue per impression. They’ll combine CTR and conversion rate into a simple number to show you which overall combination is performing the best for you.