Blog categories

search Created with Sketch.
search Created with Sketch. Search this blog

Back to blog

Get Started with Google Adwords Campaign Experiments – Part 1


Google AdWords / July 19, 2012

If you have ever managed any PPC campaigns you have at one point or another had a gut feeling that you may be bidding either too much or too little but could never prove it without fully committing to your hunch.

You may have been in a case where you tend to dominate the top of the page, have solid CTR and convert well but you have a sneaking suspicion that perhaps by lowering your bid you will continue to convert strongly while shedding some of the extra cost associated with the top spots. Or vice-versa, you may be used to paying a certain CPC and are hesitant to bump up your bids and get more visibility.

Luckily, Adwords Campaign Experiments is a feature that allows you to test the waters before jumping into the deep end.

To be clear this feature isn’t specifically new however it is hidden and can be easily over-looked by folks that are new to the game as well as the seasoned PPC manager and sometimes for different reasons.

Critical steps before setting up the experiment:

  • Define experiment: For example increase bids by 20% to see what kind of effect the increased visibility has.
  • Define pass/fail threshold: For example if you’ve increased bids by 20% and actual average CPC increases by 10% you need to have a proportionate or preferably greater increase in site engagement, conversions, sign ups, or whatever goal you have in mind.
  • Determine length of test: This will be dependent on how much traffic this test will receive and how quickly you can see the effect of these changes however you will likely want to let it run at least a week.


How to set it up:

  • Click the settings tab of your chosen campaign
  • Scroll down and open the Experiment tab
  • Click specify experiment settings

  • Name your experiment appropriately (Ie. “20% Bid Increase”)
  • Choose your traffic split (50/50 is fine for smaller or less drastic changes but adjust according to your personal comfort)
  • Choose start and end dates
  • Click save

These bid changes can be done on all types of campaigns and depending on your targeting settings you may be looking at experimental bids at the ad group, keyword, placement, topic, or audience level.

Bid Set Up:

  • Let’s take an Ad group bid increase as the example.
  • Go to the Ad groups tab
  • Click segment by experiment

  • Select all Ad groups you would like to test and click edit
  • Click CPC bid in the Experiment row and change accordingly and save

(Note** if you’d like to put the same change for all ad groups you should click the copy to all rows button that pops up beside your bid change.)

Again if this is a high traffic Ad group be sure to keep an eye on this within the first few days and of course be sure to review your findings at the end of the experiment to make a call. You can implement or delete the experimental changes in the settings area you set it up in the first place depending on your results.

If you didn’t already know about this feature you now have a new way to test out bid changes on a smaller portion of your traffic to see if it has a positive or negative effect before you fully commit.

In part 2 of this post we’ll look at how to interpret the results.

If you’ve had your own successes or failures with this feature feel free to share them, we’d love to hear about it!

Read it first:

Get every Bloom delivered fresh to you inbox

We will not spam you!


Jeff Johnson

Director of Paid Media @ Bloom


Know anyone interested in this article?

Recommend it

Help us spread the word