Anita Pinkerton

Finding Negative Keywords for Pay-Per-Click Campaigns

You Need Negative Keywords

Any company trying to glean relevant, targeted traffic from a PPC campaign needs a robust negative keyword strategy. Negative keywords are terms that may be related to your keywords, or that search engines’ algorithms may associate with your keywords, but that you really don’t want to be advertising on.

For example, if you happen to operate a chain of nursing homes, you want your ads to appear when someone searches on “nursing home,” but not necessarily when someone searches on “nursing home abuse.” In this example, you would add “abuse” as a negative keyword for the “nursing home” ad group in your campaign.

I firmly believe that any campaign without negative keywords is wasting money, unless the sole purpose of the campaign is to drive traffic, regardless of quality. I am continually surprised to find clients or other SEM firms running campaigns with clear objectives to attract only targeted traffic, yet they have limited or no negative keywords. I am particularly shocked by campaigns that have no negative keywords with everything running on broad match (gasp!). But I digress; match types are a topic for another day.

Engines Make Weird Word Associations

Search engines want to monetize as many of their searches as possible, and they want to show what they consider to be reasonably relevant PPC ads with each search. Unfortunately, it seems that the all mighty dollar speaks louder than obvious semantic connections, and engines make associations among terms that you wouldn’t necessarily arrive at on your own.

For example, in a campaign for a mini storage provider, I was running the term “storage” on broad match. I reviewed the campaign’s query report and found that Google considered “sterlite” and “rubbermaid” (both brand names of plastic storage containers) reasonable broad matches for “storage.” I get it, I get it, but it’s not at all what I wanted. I promptly added these terms as negative keywords to the ad group and adjusted the match type on the term “storage.”

Finding Negative Keywords

So how do you find negative keywords when you can’t read the engines’ minds? For all of our PPC Marketing clients, I use the following measures to generate negative keyword lists:

  1. Brainstorm
    Many times, you can easily come up with a list of obvious negative keywords. This would serve as a starter list of negative keywords for your campaign, but don’t stop here! Remember, the engines are employing their super powers to make associations that are less than intuitive.
  2. Use Keyword Suggestion Tools
    Enter your keywords into the engines’ keyword suggestion tools and review the results. This gives you a glimpse into the way the engines think. Any terms in the results that really aren’t relative for your campaign should be added as negative keywords.
  3. Use Google’s Query Report
    Though the query report does not reveal exactly how Google matched every query to keywords in your campaigns, it does hold some juicy nuggets that you can use. I check these reports for our clients at least once a month to ensure their keyword lists are as robust as possible.
  4. Search on Keywords
    Perform searches on keywords in your campaign and see what types of ads and results the engines return. This can help give you some insight into the associations the engines are making for your keywords.
  5. Analyze Web Server Logs
    If you have the hardware and manpower, the best way to track down all queries associated to keywords in your PPC campaign is to structure your log files to pull the query from the referring URL. Ahh, data! Once this method is in place, it is probably the most robust option for identifying negative keywords, but you’ll definitely need a coder and or DBA to help set up some kind of tools to make the logs palatable.

NEGATIVE KEYWORDS SAMPLE LIST

When I first launch almost any PPC campaign, I use the list below (and their plurals) as a foundation of negative terms. As you can see, these terms could be used in conjunction with almost any core term as part of a searcher’s query, but they indicate the searcher’s intent is divergent from the core term. For example, someone searching on “boat certification” probably has a different intent than someone just searching on the term “boat.”

Association
Book
Build your own
Certification
China
Class
Club
Course
DIY
Do it yourself
Dvd
Europe
Free
How to
India
Instruction
Journal
Magazine
Make your own
Manual
Newspaper
Organization
Seminar
Sex
Teacher
UK
Video
xxx

Creative fields: Opinons and News

Logged in as . logout »

Stumble It
Add to Del.icio.us

Did you like this post?

Digging and sharing is a great way to say thanks!

Leave a Reply