• http://seorecruiter.com SEORecruiter.com

    Hi Josh,

    Another great post! I believe the answer to your question is “it depends on the specifics of the industry.” As an example, candidates who come from the online ticketing industry can work with keyword combinations in the millions, where in other industries, they may only work with tens of thousands. Further, I have met average folks who work with millions of combinations, and Rock Stars who work with 25K!

    Thanks,

    SEORecruiter.com
    Michael Adams

  • sdcdog

    Hey Josh, how is creating modifiers more effective than just phrase matching a term? Won’t phrase matching “jeans” get you in front of all the variations you want?

  • doogle

    Agreed it completely depends on the industry.
    It’s also worth remembering that if you have thousands of zero impression keywords in an adgroup it can affect the quality score of other keywords in that adgroup. This is what my google rep told me but I’d be keen on hearing any other thoughts on that.

  • http://www.venturen.net ventureN

    Josh,

    Great article. I love geo targeted search segments due to the huge number of combinations. It can very easily reach into the millions. Be sure you have a handle on that kind of keyword management. Keep it tight to start, create a system, and build from there.

  • Stupidscript

    EXTREMELY CRITICAL NOTE: ALWAYS and ONLY use these types of lists as Phrase or Exact matches … NEVER use any part of them as Broad match. If you place any of the terms in the list as a Broad match it will take precedence over the Exact and Phrase matches, it will cost more, and you will receive far fewer qualified visitors. Here’s a quick illustration of what I mean:

    Keyphrase: XL blue t-shirts

    If it is set as a Broad match, then it WILL be matched with any query containing “XL” (or even “OC”, if there are few other ads that match … amazingly), any query containing “blue” (or “green” or “red” or any other color), and any query containing just the word “t-shirt” (or “soccer shirt” or “shirt tail” or “t-ball” or … yaddayadda).

    Remember that Google’s (and the other SEs’) preference is to display your ads as many times as possible, and not just for the queries you hope it will display for. Also remember that the algorithm is very, very “greedy”, and will often display your broad-matching keyphrases for queries that are barely recognizable as being syntactically-related … and I do mean *barely*.

    So, fair warning … from one who has been there and back … NEVER use ANY of the terms generated by this kind of concatenation exercise as Broad matches. Stick strictly to Exact and Phrase, watch your budget and click-throughs closely, and be prepared to act quickly to remove keyphrases that turn out to be pigs. Google won’t help you manage campaigns of that size, and believe me, it’s no walk in the park.

    On a side note, we originally used this technique back in 2003 when we generated nearly a million terms and drove Google crazy getting them into hundreds of campaigns. (Sorry about that … we may have been one of the folks who caused G to place limits on such things.) My boss said, “If they don’t click, we don’t pay … and if they click, we want to pitch to them.” After a very short period of time, it became obvious that those ad clicks were a long way from income-generating clicks, as most of them came from unqualified visitors.