Campaign experiments are a great way to maximize the value of your campaigns. Experiments are a great tool to see where your campaign could be improved and where your campaign is excelling!
How to Set up a Campaign Experiment in Google Ads
- Sign in to your Google Ads account
- From the left page menu click Drafts and Experiments
- Under Campaign Drafts hit the + button to add a new draft
- Hit the + button under Campaign Experiments and create a new experiment using the draft you just created
- Customize your experiment
- Click Save
1. Sign in to your Google Ads account
This first step probably seems pretty obvious, but we thought we’d cover it anyways. If you haven’t created a Google Ads account yet, click the Start Now button to create one. If you already have a Google Ads account, click the Sign In button on the home page.
2. Select Drafts and Experiments from the left page menu
Once you’re into your Google account, there should be a left page menu. Click on the Drafts and Experiments button down at the bottom of the menu.
3. Under Campaign Drafts hit the + button
Under Drafts and Experiments there are two sub categories titled Campaign Drafts and Campaign Experiments. Click on Campaign Drafts and then click the blue + button to create a new draft. All you need to do for this step is enter the campaign you would like to create the experiment for and then enter a name for your draft. Make sure the draft name is similar to the campaign name but is not the same so you can tell the two apart.
*Please note that shopping campaigns cannot be made into campaign experiments
4. Create Your Campaign Experiment
Once your draft is complete, click on Campaign Experiments and again hit the blue + button to create a new experiment. From here you will asked to select a draft for the experiment, name the experiment, and choose a start date. When naming the experiment you want to make sure it doesn’t share a name with the campaign you are experimenting with as this will only make it more difficult to distinguish between the two. Your start date can be the same day you created the experiment, or far off into the future. Choose your start time based off of your specific needs from the experiment.
5. Customize Your Campaign Experiment
The final step in creating a campaign experiment is to customize the budget, allocating how much will be spent and how you want your campaign to be shown when a prospective customer searching for your product.
The budget is determined by a percentage split, meaning the experiment will take the current budget allocated to the original campaign and split it between the original campaign and the campaign experiment for the duration of the experiment. The default setting is 50%, meaning half of the current campaign’s budget will stay with the original campaign and the other half will be allocated to the experiment.
Next, under Advanced Options you can choose how to split your experiment. A Cookie-Based Split means that users can only see one version of your campaign no matter how many times they search for it. This can help ensure you have the most accurate results by preventing external factors from skewing your data. A Search-Based Split lets you randomly assign users to either the experiment or the original campaign every time a search occurs. This will get you results significantly faster, but it will not get you as accurate of results because the same person could see both campaigns if they searched multiple times.
6. Click Save
Once you’ve customized the experiment to your liking click save and it will go live at the selected date. You can now see your campaign experiment under Drafts and Experiments and once it goes live you will also be able to see it under Campaigns.
Commonly Asked Questions:
What campaigns should you set up experiments for?
Campaign experiments are a great way to see if your campaign is performing at it’s absolute best. You typically want to start with successful, well-established campaign and use the experiment function to test if it could be even better with different settings.
What should my budget be?
The recommended budget split is at 50% for optimum results. If you choose to deviate from the recommended split your campaign experiment will still function just like any other experiment, your data may just take longer or be more skewed than if you were to stick with the 50% split.
How do I know if my campaign is performing well?
While what is considered “good performance” greatly varies from business to business, there are a few things to look for when you’re first getting started. A CTR (click through rate) of 2% or higher is considered a good interaction rate with your website, and anything below a 1% is very bad. When looking at conversions, the best tool to use is the Conversion Value/Cost. This calculates your ROAS (return on ad investment) and a ROAS higher than 4 is generally considered good.