Dos and Don’ts for Contextual Keyword Targeting, Part 2

Maria Breaux
|
August 19, 2013
display2

This is a guest post by Jana Fung of MixRank.

In my first post in this series, we looked at the 5 biggest don’ts when it comes to contextual keywords. Now let’s consider a few things you can do to make your ads a success:

  • Do: Experiment with lower bidding strategies. When you’re testing out campaigns, it is usually recommended to start with a large budget and optimize as you go. However, clicks on the GDN can be up to 75% cheaper than paid search clicks. So don’t be afraid to start a test campaign with a small budget. If you find that you’re continuously missing out on impressions due to budget constraints, increase your budget incrementally and optimize your keywords and placements as you go.


  • Do: Try targeting keywords that are outside your current product offering. Although counter-intuitive, this actually expands your reach to the right audience. For example, if you’re selling luggage, your target audience includes travelers and you would use travel-related keywords. However, you may also want to use the keyword “airport security rules.” Although not included in your product offering, this keyword will reach an audience that is new to travelling and in need of luggage. Targeting keywords that are outside your product offering works well because keyword targeting on the GDN is not like paid search where you try to match intent. On the GDN, keyword targeting and ads should be created to reach new audiences, and should focus less on exact match product offerings.


  • Do: Spy on your competitors’ contextual keywords and scale your own keywords accordingly, by using free competitive intelligence tools like MixRank. As mentioned in Part 1 of this blog series, you shouldn't be using a thesaurus to identify relevant keywords for your target audience. Free competitive intelligence tools like MixRank provide instant access to your competitors’ top performing contextual keywords and give you relevant keyword suggestions that you can test with your own campaigns.


  • Do: Add other layers of targeting in conjunction with keyword targeting. In addition to keyword targeting, the GDN offers placement targeting, interest targeting, topic targeting and re-marketing. With the exception of re-marketing, you should add at least one layer of targeting to increase precision for your intended audience. Keep in mind that if you add too many layers of targeting, volume may significantly decrease, so be sure to check the estimated display reach within your Display Network tab.


  • Do: Limit the number of keywords per ad group to 3-6 to increase precision, relevancy and Quality Score (QS). Because keyword targeting on the GDN only allows for broad match, it doesn't make sense to add variations of the same keyword as you would in paid search for exact and phrase match targeting. If you’re targeting too many broad match keywords that aren't thematically related within a single ad group, your QS may suffer and hurt campaign performance.


While this is just a starter list of what you should consider when building out your contextual keywords and ad campaigns, it is imperative to understand the differences between the Search network and the Display network. Display advertisers are creating ads and keyword targets to reach specific audiences, while search advertisers are creating ads and keyword targets to reach specific intentions.

What contextual keyword dos or don’ts have you had success with in the past? Share them with us in the comments section below.

About the Author
Jana Fung, guest author of this post, is the Marketing Manager of MixRank. She has managed successful demand generation programs for over 6 years. MixRank.com is a spy tool for contextual and display ads. With MixRank you can see exactly where your competitors are buying traffic and which ad copy is performing best for them across over 100,000 sites. If you’re a MixRank fan or just want to say hi, Jana is interested in connecting with you! Follow her on Twitter @jana_fung.

Sign Up and Get Updates by Email
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.