UpriseUp - Up
UpriseUp - Rise
UpriseUp - Up
Back to EventsBack to Blog

‘I call it Experiment 626!’ How to Set Up Google Ads Experiments

Google Ads Experiments Blog Lead Image

How to use Google Ads Experiments

Adjusting and optimising your campaigns is key to getting the most out of Google Ads, but what if you’re not sure whether a change will be in your best interests or not? In Paid Media, you always want to ensure you are getting the most benefit out of your Ad spend. This is where Google Ads Experiments comes in. At Uprise Up, we’re always looking to continuously improve the accounts we work on and as a data driven agency, experiments allow us to learn directly from the data produced by these.

What are Google Ads experiments?

Experiments are a way to test a change that you’re not certain about in your Google Ads account. Whatever change you are debating, experiments are a way to test your theories without taking any major risks. For charities in particular, this is ideal in keeping costs down and not needing to spend any more than necessary in ensuring your campaigns are running the best they can. Experiments can also be used in either Google Ad Grant or paid accounts, meaning that charities only running Ad Grant activity can also benefit from these.

This change could be from a range of areas, such as:

• A change in bid strategy
• Using pinning in your responsive search ads
• Changes in your ad copy
• Using dynamic keyword/location insertion headlines .
• Using dynamic search ads
• Different landing pages
• Pausing/adding ad groups
• Different keyword lists
• Applying negative ad targets
• Audience or locational targeting

Google Ads uses A/B testing, also referred to as split testing, that allows multiple variants to run simultaneously to show which performs better. This uses a ‘control’ campaign (A) against a ‘variation’ campaign (B) to see what, if any, difference in performance arises between the two campaigns. All you need to start an experiment is an existing campaign and an idea of what it is you’d like to test.

Before you start: Experiment limitations

It’s worth noting that there are some limitations to which campaigns are allowed to use experiments. There are only a few restrictions but it’s important to be aware of them before you start, otherwise Google might not inform you until you’ve already put in most of the work in setting one up.

Restrictions include:

  • Campaigns in a shared budget. Removing the campaign from the budget while the experiment is carried out is a route around this, but no experiment can be run if a shared budget is still attached.
  • Campaigns that include ‘Text Ads’, even if they’ve been removed. This is referring solely to the old style of ad referred to as ‘Text Ads’, rather than ‘Expanded Text Ads’ the latter of these function in experiments just fine. What’s unusual here is that this restriction applies even if the ads have been ‘removed’ from the account, so if these are in a campaign even in this state, you won’t be able to run an experiment.
  • Custom experiments are not available for App of Shopping campaigns. Experiments are only available for Search, Display, Video and Hotel Ads campaigns.
  • Only one experiment for a campaign at a time. However, you can schedule up to five experiments for a single campaign.

 

How to set up Google Ads Experiments

Let’s run through an example. You’re running a campaign focused on recruitment and volunteering and you want more application-based conversions for a campaign, but you’re not sure what effect a bid strategy change might have on your campaign. You decide to set up an experiment to test this and have directly comparable data between your original campaign and your variation.

Here’s the set-up process:

Creating a custom experiment - how to set up Google Ads Experiments

First of all, select the ‘Experiments’ tab in the left navigation bar, while on the ‘All campaigns’ view as shown above. From here, we’ll make a custom experiment, allowing us to test as much or little as we like.

Select the campaign you want to make into an experiment in Google Ads

Next, choose the campaign you want to make an experiment for and name it something clear and to the point, such as stating the purpose of the experiment.

Changing the settings for Google Ads Experiments

Now, you can make your changes. Go into the campaign settings and change the bid strategy over to our desired option, just the same as you would to make any changes outside of an experiment.

Setting a Google Ads Experiment live

Then you can simply set it live! While you can choose the metrics you’re most interested in to see more directly in your results, you’ll still get a range of metrics to view once your experiment is running. Additionally, if you have a larger campaign or limited budget and only want to test this change on a smaller level, here you can set just how much budget you want to split between the base campaign and the experiment.

Generally speaking, these experiments do need some time to learn, so we’d usually suggest running an experiment for at least 4 weeks, otherwise this simply won’t give enough time to see the effects of any changes, although this can depend on just how much data is coming through.

It’s also important to note that while you can extend the duration of the experiment once it’s begun or choose to end it early, you can’t reenable an experiment once it’s ended, so always double check that you’ve given the experiment enough time to test your changes effectively.

Another feature that’s incredibly useful here is Enable Sync. For most experiments, this allows us to continue optimising across both the control and experiment campaigns as we go, such as adding new keywords or adjusting copy. A case where you might not use this is in an experiment where copy is being tested, as you might not want any further tweaks in your copy to pull across to the experiment.

Monitor performance in Google Ads

As the experiment is running and learning, you’ll want to monitor performance. We would recommend regularly popping back into the experiments tab throughout the duration of the experiment, where you can easily take a look at the performance of your experiment compared to the base campaign.

Summary data interface for Google Ads Experiments

The summary, as shown above, will highlight whichever metrics you decided to look at earlier and will also display a range of metrics to compare. This can be customised in the same way as you’d customise columns within the regular campaigns view.

You can also make any alterations to your experiment here (although we’d recommend keeping an experiment the same from start to finish), apply the experiment early or end the experiment early if you’d like to. Once the experiment is over, you can use this summary to get a clear idea of your results and then decide whether or not to apply your changes to the base campaign.
And that’s all there is to it! Experiments are a great and adaptable way to safely test your ideas out and directly see just how your changes might affect a campaign.

Need some support?

Do you want some support setting up your Google Ads and running experiments? Or would you just like to talk about your paid media in general? We’d love to hear from you! Check out our contact us page or email us at hello@upriseup.co.uk, or simply send us a message through our contact page.

    Did you enjoy this blog post?

    Share this article:

    We use Mailchimp as our marketing platform. By clicking below to subscribe, you acknowledge that your information will be transferred to Mailchimp for processing. Learn more about Mailchimp's privacy practices here.

    Contact us