AdWords
1.7K members online now
1.7K members online now
Use AdWords conversion tracking and reporting to measure your results. Have a question about Google Analytics? Post it here, on the Google Analytics Community!
Guide Me
star_border
Reply

Should I include search partners choosing a winner in a splittest?

Follower ✭ ☆ ☆
# 1
Follower ✭ ☆ ☆

I recently changed all campaigns to include only search netword, inclusive search partners. Search partners deliver a lower CTR and are performing worse on all parameters. I am wondering if I should include the stats from search partners when choosing a winner from an ad splittest?

 

Google search accounts for 90% of the clicks, so I am thinking I should only focus on optimizing for this network only. What are your takes?

1 Expert replyverified_user

Re: Should I include search partners choosing a winner in a splittest?

Badged Google Partner
# 2
Badged Google Partner
Opt out of search partners if it's not working for you, and taking budget away from more profitable areas.

Tom

Re: Should I include search partners choosing a winner in a splittest?

Follower ✭ ☆ ☆
# 3
Follower ✭ ☆ ☆
@tomhalejr

search partners are giving ok result but account for very little of the total clicks. Therefore I want to continue advertising with search partners, but I am wondering If i should include the numbers from search partners when deciding on a winner of my split test? Some times search partners give conflicting results!

Re: Should I include search partners choosing a winner in a splittest?

Badged Google Partner
# 4
Badged Google Partner
If you are talking about the actual ads within the adgroup(s) of the campaign, then yes.

Since the network settings are at the campaign level, and you can't target search partners only, you will have to make compromises. If one ad works better for search partners than another, and search partners is the most profitable network, then use the best search partners ad as your control.
Tom

Re: Should I include search partners choosing a winner in a splittest?

Follower ✭ ☆ ☆
# 5
Follower ✭ ☆ ☆
and if Google search accounts for 95% of the clicks/revenue than I would not include search partners when deciding on a winning ad?

Re: Should I include search partners choosing a winner in a splittest?

Badged Google Partner
# 6
Badged Google Partner
Of course I would include 95% of the data in any test. Smiley Happy

Since network settings are a campaign level setting, I analyze the network performance at the campaign level. If search partners isn't working for that campaign, then opt out.

When it comes to (search networks) ad testing, it's all about the user experience/expectations in relation to the search query, ad message, and landing page. You can't isolate networks at that level of the account.
Tom

Re: Should I include search partners choosing a winner in a splittest?

Follower ✭ ☆ ☆
# 7
Follower ✭ ☆ ☆
The question is not if I should include the 95% - it is if I should include the 5%, bearing in mind that sometimes it contradicts the 95% Smiley Happy

Re: Should I include search partners choosing a winner in a splittest?

Top Contributor
# 8
Top Contributor
Hi Kleland,

I'm going to disagree with my friend Tom here. That doesn't happen often. You can take this for what it's worth.

You can effectively create a search partners only campaign. Clicks on the search network are generally much less expensive than on Google itself. You have some pretty good data on which to base your bids.

First, segment your data by network. It seems you've already done that. You will see the CPC's of Google and partners separately. You will no doubt notice there is a big difference in the numbers. What you do create a Google only campaign, with bids high enough to get impressions on Google. You can figure that out bey looking at the segmented data. Then, create a campaign including Search Partners, but bid only enough to get impressions on partners but not high enough to get impressions on Google.

I would suggest you keep the current campaign and set it for Google Search only. Then duplicate the campaign, include Search Partners, but with significantly lower bids. You will effectively have two campaigns, one for Google search and one for Search Partners.

The only drawback to this solution is you now have separate budgets to manage for each network. That could turn out to be an advantage.

After splitting the campaigns like this, check the segmented reports frequently to make sure the Search Partners campaign is not "stealing" impressions from the Google campaign. You may need to adjust your bids, either the default bid or the keyword level bids.

After doing this you'll be able to pick different Ads as winners for each network, and all the data within the AdGroup will be applicable.

Best of Luck!

Pete
petebardo -- Deadhead doing AdWords

Re: Should I include search partners choosing a winner in a splittest?

Badged Google Partner
# 9
Badged Google Partner
I don't think we disagree at all Pete. Smiley Happy

Try looking at the campaign/adgroup level top vs. other segment, with the impression share columns enabled Klelund. With Pete's insights in mind, does that help explain why the performance or data may be different between networks?
Tom

Re: Should I include search partners choosing a winner in a splittest?

Top Contributor
# 10
Top Contributor
Hey Tom,

I'm only disagreeing on creating a Search Partners only campaign. Other than that, great answers for Kleland.

Pete
petebardo -- Deadhead doing AdWords