Not a new concept by any means, but more and more accounts aren’t using it enough. With the introduction of modified broad match advertisers should consider using keyword lists from one ad group or campaign to filter and “push” traffic from one ad group to another. In order for this work though, you’ve got to have a solid account structure in place that is segmented and firm.
Still wondering what the heck I’m talking about?
No problem because I am going to walk you through a real life example…
Client A sells an online math game for kids for grades kindergarten through 3rd grade. Their top keywords are grade specific like “first grade math games” “first grade math” “1st grade math games” and “1st grade math.” These 4 terms and the same style for the other grades (kindergarten, second and third) are the ones that consistently perform in conversions over and over.
Yet at the same time, there are some longtail keywords and lesser searched keywords that I also want to test and bring conversions but I don’t want to disturb my money makers. Sure, I could do phrase or exact match on those and I did try it but what ended up happening was choking off more traffic than I wanted to. I ran search query reports every week, but that’s a reactive and slow process and since I already have a substantial negative keyword list in place, how do I really keep that traffic flowing?
One campaign for each grade, one ad group for the top 4 terms, one ad group for my longtail and experiment keywords. A top level negative keyword list at the campaign level keeps out the freebies, unrelated and other subjects (like reading or software programs) and the other grades to keep from first grade ads from showing up with second or third grade queries.
And the real meat of the control- adding exact negative keywords at the ad group level.
Example: first grade
Campaign level: free, download, software, reading, third, second, fourth, third grade, second grade, etc;
Ad group- First Grade Longtail Keywords: [first grade] [first grade math] [1st grade math] [1st grade math game] etc;

Click for full size display.

You might wonder why I didn’t create a 3rd ad group or new ad group for the experimental terms to separate them out from the longtail keywords even further- the answer is traffic. In this case, the volume simply isn’t there and neither is the time or necessity to manage it to that level. Plus with the addition of the Experiments feature, there really is no need to- I simply use that new wonderful tool.
The ad groups and campaigns themselves are highly relevant, dialed in and structured in a way that Quality Score does not become an issue.
Each grade had their own grade specific landing page with the top 4 converting keywords on it and since the secondary ad group’s keywords all had relative terms (just variations of really) they weren’t hurting each other. Quality Scores for each keyword in these ad groups never went below a 7 and the top keywords enjoyed 10s.
By creating a “fence” around my top terms and secondary terms, I am forcing Google to show the ads that I want with the keywords that I want while still garnering those relevant fringe queries. I know what ad message works to get those top 4 keyword producing and I have a set of those ads to continually bring in that stream so it equals little maintenance for me. I can now spend my time on the secondary terms and looking to expand those 4 or trying alternative messages, destination URLs or bids.
This approach is particularly useful if you want to cut down on the amount of time you spend managing your account and do not have something like a 3rd party paid search management platform. Even more so if you’ve got a really tight budget or seasonality where you want minimal maintenance and can simply drop the budgets or pause ad groups to keep everything really under control. It is especially useful if you’re in a niche industry like Client A is.
Still run those search query reports though…