• http://twitter.com/harryfassett harryfassett

    QS is like you said Jenny, comparing it to the old Google PR, which is pretty visual but what does it mean in terms of results, and usually it’s very ambiguous at best, so it’s neither here nor there as the old saying goes, and won’t be missed for it’s added confusion. :) day1charitydonation

  • http://www.geekpoweredstudios.com Guillermo Ortiz

    I’ve seen a lot more 10’s lately from keywords that I’ve been managing for a long time so I’m going to wager that those were earned ;). I will have to keep an eye out for 10 scores from newly minted keywords that have yet to be tested. Thanks for the heads up!

  • http://saidulhassan.com/ Saidul Hassan

    Re:Random Quality Scores Of 10… I’ve been testing this for some time now and now I’m starting to drop some to test the avg. QS for ad groups. My hunch is they do influence.

  • http://twitter.com/echwa Damien Anderson

    When you split out ad groups for different match types, do you use broad or modified broad?

  • http://twitter.com/jennyhalasz Jenny Halasz

    always modified broad! Wish Google had that as a standard match type.

  • Guest

    Haha, now if only I could get my people to stop focusing on it. ;-)

  • http://twitter.com/jennyhalasz Jenny Halasz

    It is an interesting analogy to PR, isn’t it? I’ve always wondered if it will someday disappear.

  • http://twitter.com/tedives Ted Ives

    Since QS is largely CTR (see Hal Varian’s video on Quality Score), focusing on it is really only helping Google. Google wants to sell clicks, so of course Google wants account managers to work on achieving higher CTRs.

    If you’re bidding on the term [britney spears] and all you’re selling is britney spears calendars – you *want* a really low CTR or you’ll lose a ton of money. So your creatives in that case need to be designed to *discourage* general searchers from clicking – the opposite of the marketer’s tendency to try to shoot for as high a CTR on ads as possible.

    I think your QS distribution can tell you if you have some sort of sitewide problem – for instance, if your Quality Scores are all capped out at 7, with no 8/9/10s. For instance – I’ve had several clients who had no privacy policy/TOS on their websites whose Adwords accounts had that problem – they created Privacy Policy and TOS pages and added footer links to them on every page, and a few weeks later all their 7’s magically became 10’s.

    But otherwise I don’t think QS is much worth agonizing over.

    BTW – *very* interesting table on not set vs. not provided – thanks!

  • Tally Keller

    I’m not sure why you would bid on a term if you don’t want people to click on the ad.

  • Tally Keller

    I still think QS is important to benchmark over time, but, like any metric, you can’t obsess over every tick. Take it with a grain of salt – if cost per acquisition is improving, then who cares about QS?

  • cjvannette

    Can you elaborate on “creating multiple copies of each ad group for all of the match types we want to target”? I’m a newbie and I’m not sure what you mean.

  • http://twitter.com/jennyhalasz Jenny Halasz

    Thanks for your very insightful comments! I hadn’t really thought about the effect CTR has in cases where you want the CTR to be low and targeted. Interesting! I also love the tip of TOS/PP pages. You are absolutely right; I’ve seen that work as well.

  • http://twitter.com/jennyhalasz Jenny Halasz

    Sometimes you want to bid on a more general term and use the ad creative to narrow the people who click through. In the example above, you might bid on “Britney Spears”, but make your ad copy something like “Calendars of Britney Spears, We only sell calendars.” That’s pretty lousy ad copy, but hopefully it gets the point across. Of course, for a keyword like “Britney Spears”, it’s unlikely that having such a low CTR would get you a spot in the auction.

  • Pat Grady

    GREAT run down!

  • http://twitter.com/jennyhalasz Jenny Halasz

    I’ve been asked this several times, so I guess I’ll give out my secret sauce… we do one ad group on modified broad, but add all the exact matches as negatives. Then we do one ad group on exact match for the same keywords.

  • http://twitter.com/jennyhalasz Jenny Halasz


  • http://twitter.com/jennyhalasz Jenny Halasz

    Very true! Still, as a marketer with roots in SEO, I can’t help but analyze the small stuff.

  • http://twitter.com/jennyhalasz Jenny Halasz

    Christine, with all due respect, I would *never* recommend putting all of your keywords into one ad group. That’s a very bad idea, unless you only have 15-20 keywords total.

  • http://twitter.com/jennyhalasz Jenny Halasz

    Glad to hear that, Christine, and sorry for the confusion. Still, we do separate the ad groups so that we can apply different max daily budgets to them.

  • Lloyd Cohagen

    Re:QS – I find sometimes that my highest ROAS keywords have a low QS based on low CTR. I assume it’s due to impression

  • Brad Seraphin

    I wonder how the economy of people being paid to improve QS compares to the economy of paid search. Moreover, what % of the entire search budget is spent on chasing this unicorn? I suppose I ought to thank google for creating this matrix in which we can all get paid to wax knowledgeable to different degrees on a subject that seems to be little more than common sense once you decode the industry jargon.

  • http://www.facebook.com/DallasPPCGuru Kevin Adams

    In my previous tests I’ve noticed in a lot of cases that the www actually improved QS when the ad appeared above the organic results on a desktop, but the opposite was the case on mobile or on the right-hand side. Is anyone seeing the www being a drain on CTR?

  • Steve Bookspan

    Great info about about Adwords. I have noticed QS all over the palce as well. I am anxious to see what else you find out.