We’ve had a lot of feedback on additional ways that you can segment exactly what ads get included in a multi-ad group test and we’ve just launched new options for determining what ads are included:
When you define a multi-ad group test, you must select the campaigns to be included; which is the same way its been in the past.
Optionally, you can choose to only include ad groups if they match (or don’t match) ad group labels. If you have ad groups that you don’t want to include in a test, you can label them and then choose ‘adgroup labels contain none’ and those ad groups won’t be included in the test. Alternately, if you only want to include certain ad groups in a test, you can label those ad groups and then chose only to include those ad groups.
Choosing ad group labels is not required, but if you want to only include a subset of ad groups in your test, you can now do so with this new selection criteria.
The use of wildcards in multi-ad group patterns has been widely used to test:
What you couldn’t do was test if a line didn’t include something, such as your brand versus a non-branded ad or DKI ads versus non-DKI ads, etc.
We’ve just rolled out several new pattern options you can use in defining ad tests:
| Pattern | How ads are matched to the pattern |
| abc | abc exists anywhere in the text |
| * abc | abc exists at the end of the text |
| abc * | abc exists at the beginning of the text |
| abc * xyz | abc exists before the wildcard and xyz exists after the wildcard |
| abc * xyz * | abc exists at beginning. xyz anywhere after abc |
| * abc * xyz | abc exists anywhere before xyz. xyz exists at end |
| !!abc | abc doesn’t exist |
For instance, if you wanted to test how DKI does against non-DKI you can set up this test:
That test says for pattern 1; if the ad does not have {keyword: in it; then include it in the test result for pattern 1 and then pattern 2 is testing ads that start with dynamic keyword insertion.
This would let you see if dynamic keyword insertion is better or worse than your statically written ad.
As more people are testing ad customizers, dynamic keyword insertion, DSAs, and other new ad formatting options; we’re ensuring that you can easily test these parameters either against each other or just test if using these formatting option are actually better than your static ads. We’ve seen cases where ad customizers lowered the account’s performance; so it’s a good idea to test if taking the extra work to customize ads is helping your bottom line.
Please note that these new operators also exists for the new ad and ad group filters. It’s now easy to find all ads that don’t match a certain brand message and pause them all, or ads that have an old code or promotional message and replace all the text in them at once.
We’re working towards helping you be consistent in your brand messages, change old messages in bulk, and pushing the limits on how you can manage your ads. We have some new features in the works that will make some of these tasks more robust and even easier.
Related new features:
If you have any feedback or questions on the new features, please let us know.
