How to Improve Low Quality Score Keywords: A Case Study

By , Senior Account Manager at Hanapin Marketing

253 SHARES

Google is judging your keywords for quality, judging them hard.

There are a lot of ways to improve low quality score keywords, and with Google’s new improvements to quality score, they’re making it a little more transparent as to what you need to do to improve your low quality score keywords. I’ve found in my accounts that it’s most often ad relevancy and/or expected click-through rate. Since these are the easy, faster fixes–they’re the ones I focused on first. Landing page fixes can take up to six weeks for Google to recognize. You should still do those, and we have a hilarious little video blog on how to do that, but I wanted to share with you today something I did in one of my accounts to increase quality score through ad relevancy and CTR.

1.) Identify your most likely to succeed low quality score keywords.

I just downloaded a keyword report and filtered out any quality scores higher than 2/10. From there, I looked through to see if any had converted. Those are good places to start. Otherwise, though, just look through your keywords at ones you think should work. So, let’s say you sell shoes, and one of your low quality score keywords is “black and white running shoes”, and you not only sell those, but have a landing page for them. That’d be a great keyword to pick to target. I also want to suggest that you only use exact or maybe phrase match at first. Especially if you’re picking general short-tail keywords. If you take something like “shoes” that was limited by a low quality score at broad match and take it even to a 3/10, you’re really opening up the flood gates, and you could spend a lot really quickly. You may or may not see the return on that keyword. So, for me, I only used exact match for this.

2.) Make a whole new campaign for your low QS keywords.

This is not only going to make it easier to manage your breakouts, but it’s going to also ensure any campaign level quality scores won’t affect your new campaign. Just the account wide quality score. I called my campaign “Low Quality Score Breakouts”.

3.) Breakout your ad groups in the most segmented way possible.

The point of this whole thing to make the most targeted ads possible, so even if it means having an ad group for each word, doooo eet! So, put “black and white running shoes” in a separate campaign from “red running shoes”. Just think: if I’m writing an ad for this ad group, can I write one with a keyword in it and it’ll make sense to people searching for the other keywords? You could have “black and white running shoes” “white and black running shoes” “running shoes in black and white”, etc. all in the same ad group, for example.

4.) Write two ads to test against each other for each ad group. The most targeted, relevant ads you’ve ever written in your little ‘ol PPC life. Here’s a good one for the example we’ve been using:

If you need more help writing ads, check out this webinar the lovely miss Kayla Kurtz did a couple of months ago on ad writing and testing.

5.) Set up test budgets and push those suckers live!!

I have a budget I have to stay in every month, so I did some mathing (that’s a verb, now) to figure out how much I want to spend on these keywords and this campaign. You want something that’s going to allow your keywords to get enough clicks to convert, but something that’s not putting your account at risk of tanking metrics for the month. If you need some help with this, check out our blog article about bookending for new keywords.

6.) Check your results.

Drumroll, guys…the SAME DAY I did this, one of my 2/10 keywords went to a 4/10 and converted twice at 20% of our goal cost per conversion.

 

Keyword before breakout:

Keyword AFTER breakout:

 

So, that’s what I’ve got for you, folks. And I think it’s pretty spectacular. Let me know if you have any similar stories, suggestions for things I should do differently, or fun comments!

ACO_endad_Knockout

Twitter Facebook LinkedIn Google+ Email Print More
  • http://twitter.com/pedromartheyn Pedro Martheyn

    What is the quality score right now? I don’t think that your QS still 4. When you launch campaigns Google sort of gives you a “standard” quality score base on various factors. However as the time progresses your quality score is changes according to your performance, so it might go back to 2 (or increase in many cases). In addition to this your imps are under 1,064 (Google’s magic number) which means the stats on the factors that affect quality score directly are not completely accurate. You need over 1,064 imp to have 97% accuracy on your stats. Also, you have to take into consideration match types, each match type have different QS, so if you move keywords you will ultimately end up splitting the traffic which can affect negatively your campaigns.
      

    • Amanda West-Bookwalter

       Hi Pedro,

      Thanks so much for reading & commenting!

      You’re right! When you upload the keywords into the new campaign, you’ll start with whatever your account QS is. If you have an unhealthy account, it’ll probably start low. In this keyword’s case, it started at a 3 when I first uploaded it. The stats that affect QS as far as CTR and ad relevancy go are updated all the time, so even before or after 1,064 impressions, you’ll be able to change the QS of your keyword if you affect CTR or ad relevancy. And you’re right about match types, too! That’s why I recommended using exact match, because it has the most likely chance of having higher QS, or at least as far as I’ve seen in my accounts. However, I don’t really know what you mean about splitting traffic if you move keywords. Could you elaborate on that? I would recommend pausing the keyword wherever you’re moving it from, is that what you meant? I may have left that important part out because it seemed so obvious to me!

      • http://twitter.com/pedromartheyn Pedro Martheyn

        Hi Amanda,

        Thanks for responding to my comment!

        The stats that affect QS directly go beyond CTR and relevancy as you know. As mentioned before you do need the minimum of the 1,064 to have a pretty clear idea of what your “real” quality score is for those keywords. I would say that not too many people know that Google take a look at 17 different factors (Maybe more) to calculate quality score. It goes from account structure to overall site experience (besides landing page), that’s precisely why you need that many impressions minimum. I know there is a certain amount of clicks that you need too but I just can’t recall it at this time.

        Regarding match types, you are right. Quality score on exact match tend to be higher than all the other types, except for modify broad, they seem to be equal lately. Now, what I meant by splitting the traffic is that if you have same keyword in all match types and let’s say you take your broad match and put it into a different ad group due to its low quality score, you might be end up affecting the overall performance of the campaign. Theoretically queries should get match to the “exact match” more often; however as we all know that’s not always the case. So by taking that keyword and put it into a different ad group, queries that should’ve getting matched to the “exact match” keyword will end up getting match to that broad keyword therefore your overall cost will increase. But if you have all the keywords in the same ad group, you will increase your chances to get that query matched to the “exact match” keyword, helping decrease overall cost and increase overall performance. Now what I would recommend to do would be to individually change the bids slightly between the keywords. For example the ones in exact match you can bit $1, the ones in modify broad $0.99, the ones in phrase match $0.98 and the ones in broad match $0.97 – That will give you better results and overtime your low QS keywords will feed from the high QS keywords helping them to increase overall QS. Remember there are QS at the account, ad group and campaign level. All this is taking into consideration when calculating overall QS.

        Cheers!!

        p-

        • Amanda West-Bookwalter

           Pedro, I love having these conversations! Thanks so much for getting back to me.

          Yes, I think there’s a lot to quality score that most ppc managers don’t know about, and some that no ppc managers know about that’s still shielded behind the Google curtain. We actually wrote a pretty comprehensive guide to quality score a while back: http://www.ppchero.com/ultimate-guide-to-adwords-quality-score/

          For the traffic splitting, I always think best practice is to split match types into their own ad groups and negatively embed the exact and phrase matches to make sure your traffic is getting sent to the most appropriate keyword. And, I totally agree that the strategy of bidding highest on exact and lowest on broad makes the most sense, as exact will usually have the highest conversion rates.

          For keywords that are similar in different ad groups, Google will pick which keyword to use pretty much how it does with keywords in the same ad group if you’re not negatively embedded, meaning it picks the keyword that matches the most closely, with the most restrictive match type, and with the highest ad rank. Obviously, like you said, Google isn’t perfect, which is why I would suggest if you’re pulling out an exact match type, to negatively embed it in the ad group with phrase and broad if you have them, and you’re not pulling them out too.

          Of course, this is all just a difference in strategy, and I think the most important thing is the results you see in your account. I’m happy to say it’s been 3 days, my keyword has 1,952 impressions, and it’s still a 4/10 when it was a 2/10, and it’s converting at an 18% higher conversion rate than the account as a whole for the same time period.

  • Steamer

    FWIW, quality score is only determined using exact match keywords (or exact matches if now exact terms exist). It’s a myth that different match types impact QS.

  • http://www.facebook.com/raju.paliwaal Raju Paliwal

    my account is fully optimized still my quality score is only 3.
    it is showing that my ad relevancy is below average on my mail keyword :/
    Can you please share some solution for this :) i started my campaign 2 day back.