10 Ways To Increase Your Adwords Quality Score - A Mini Case Study
So what is quality score? According to Google:
“Quality Score is the basis for measuring the quality and relevance of your ads and determining your minimum CPC bid for Google and the search network. This score is determined by your keyword’s click through rate (CTR) on Google, and the relevance of your ad text, keyword, and landing page.”
Quality score effects your keywords minimum cost per click and position in the search results and also if your ad will even show at all!
This is fundamentally the same for the other first tier Pay Per Click networks (Yahoo Search Marketing and MSN AdCenter). There are many other factors involved, however this is the most transparency we’re going to get from them at this time. For now, we will deal with the Google Adwords quality score.
Around the same time of the release of Yahoo! Panama, Google began showing keyword quality information at the keyword level in Adwords accounts. Google shows weather your score is Great, OK or Poor. It may not be much information, but it has been a great help for those trying to diagnose why their PPC bids are so high.
Last month, we had a new client who came to us for advice on lowering their minimum cost per click and ultimately increase their ROI. Our client was already experiencing a positive ROI using Adwords because of their products profit margin, however since Google started showing QS, they realized they were not doing as well as they thought. Their campaign was relatively small and they were paying an average minimum CPC of $0.40 across their three ad groups. Each Ad Group had 100 keywords in each. They also had a few disabled keywords. When we examined their account, we could see that almost 90% of their keywords had a “Poor” quality score.
Lets say our client sold “Blue Widgets”. Here’s what we did.
Campaign Account Factors
************************
1) Split Keywords into Smaller more targeted Ad Groups
We used the in built keyword grouper tool in Adwords editor to group keywords into 15 groups of 20 related keywords. eg:
buy blue widgets
exchange blue widgets
what is a blue widget
buy blue widgets online
2) Create relevant ad copy for each group
We then created an ad creative for each keyword group using the common grouping denominator “Blue Widgets”
Note the upper case on each word and use of the keyword in the title, ad copy and display URL. ALL 4 ads contained blue widgets in the same form as you see above. It should also be noted that the display URL does not need to match the ACTUAL URL exactly.
3) Optimize Creatives
We created 4 different ads for each group using different verbs and calls to action. We also turned ad serving optimization OFF so we could accurately split test all 4 ads. After a week, we chose the best performing creative and deleted the others. We then tried to increase the CTR by working on variations of that creative. After two weeks, it was plain to see which was the better performing ad. Within two weeks the overall campaign click through rate was up. Note that optimizing creatives is a never ending process. You can ALWAYS squeeze an extra 0.01% CTR by being…creative.
(This was not the creative used for the campaign)
4) Experiment With Matching Options
Our client was only using broad match for their keywords in their campaign. We added exact match and phrase match keywords to each ad group and chose which of the three had a better QS and lower minimum CPC and deleted the other two matching options. In the majority of cases, exact match won.
On Page Factors
****************
While we were working on the PPC account side of things, many of the on page factors were being worked on at the same time.
5) Link Building And SEO
We ran a manual deep link building campaign using the highest performing keywords (Volume AND Conversion Rate). Not only does this help Quality Score but also helps with the organic SEO campaign in the coming months. We submitted a Google sitemap, made the site semantically coded and corrected some navigational issues.
6) Implement Keywords
For each page we implemented most of the keywords into the copy.
7) Split Test Landing Page
We set up a multi variant test to split test the landing pages and test conversion rates. Google Analytics has an excellent write up on the major elements that can effect conversions. We cannot stress the importance of multi variant testing enough. Testing and optimizing conversion rates on landing pages is too big a subject to go into in this post. It should be noted that we also used heat maps to test the landing pages.
8) Meta Tags
We added the best performing keywords to the meta tags on each page. We also used the EXACT ad text from the best performing creative in the meta description. We also used the BEST PERFORMING and MOST DESCRIPTIVE keyword as the title tag.
9) Essential Site Pages
Although the site had a privacy policy, it was not linked in the navigation (Header OR Footer). We added the privacy policy as well as creating and adding an informative “about us” page, a “terms and conditions” page, a newsletter page and a new contact us page.
10) Make Sure Google Thinks You’re Relevant
We used the Site-Related Keywords tool to make sure that Google thought the landing page was related to the keywords we are targeting. After a little “fleshing out”, Google saw each landing page as VERY related to the keywords we were using.
Clearly for privacy reasons we have left out many details regarding our clients campaign. We increased the campaign daily budget to twice what it was and left the campaign running for the month to get more accurate figures.
The Results:
Within four days the Quality Score for 60% of the keywords went from ” Poor”to “Great” with average minimum CPC prices down from 40 cents to 4 cents.
All keywords that still had a poor quality score after the month (8 of them) were deleted.
Overall campaign click through rate was up 12%
We increased conversion rates from 3% to a whopping 24%
Our client was 100% happier.
Our client was able to slash their monthly spending budget and MASSIVELY increase their sales.
Conclusion.
Increasing you quality score and optimizing your Adwords PPC campaign is an absolute MUST if you want to reduce your spend and at the same time increase your sales. There are MANY other factors, known and unknown that go into improving quality scores, however the most fundamental have been covered. Most other PPC networks also use a similar QS ranking method and the areas I have covered can be applied to them too.
I hope this has helped anyone starting their first PPC campaign or anyone looking to improve their ROI by increasing their Quality Score. If you would like more information or would like to avail of our Google Adwords Management service, please get in touch.
March 8th, 2007 at 9:03 pm
Excellent write up Dave. A lot of effort behind that post and some really good tips on improving QS - cheers.
March 9th, 2007 at 3:52 pm
You’ve hit the nail squarely on the head here, Dave. It’s all about relevance, and from splitting ads up into smaller groups based on tightly similar keywords to experimenting with matching options, what you’ve mentioned here is spot-on in my experience.
Another tip for you and your readers — MindValley Labs has a tool that allows you to A/B test Adwords ads. You might want to check into this and test it out.
March 9th, 2007 at 5:09 pm
brilliant post Dave. Seems many advertisers were hit pretty hard with the last quality score update(including ourselves) and these tips will help overcome that a lot.
Thanks for posting!!!
March 9th, 2007 at 6:04 pm
Thanks for stopping by Rick. MindValley Labs have some AMAZING stuff over there. We use their clickmuse Adwords Optimizer alerts tool. Some of their creative testing results are amazing and spot on. They have helped us so much. We will be looking into trying some more of their tools shortly.
March 9th, 2007 at 6:10 pm
Excellent post. Is it correct that landing page quality doesn’t impact ad positioning but does impact the minimum CPC? I thought that was the case at one point, but I am not sure if that has changed.
Also, it seems that the landing page assessment makes assumptions about what will lead to good conversion (through relevancy). Is anyone aware of plans for Google to use actual conversion data (for those have Google tracking in place) in the quality score?
Thanks for an insightful post.
March 9th, 2007 at 10:39 pm
Superb post. But, i have a question. Do u write different pages with different title, description and content for each ad group?
March 10th, 2007 at 1:55 pm
Really excellent post Dave - I appreciate you sharing so many details of the methods you found successful.
March 10th, 2007 at 5:18 pm
@Tim, I use Google conversion tracking and it is satisfactory for our needs, but we are looking into clicktracks. Diorex has a nice little rant over about Google wanting conversion data.
I have to admit, he makes a very good point.
@Owl. It completely depends on the “Meaning” of the keywords and what the quality score is. It is MUCH easier to get a decent quality score if all the keywords in an ad group are on the actual landing page, but it seems more weight is given to keywords being in the actual ad.
@Janine. Thanks. This information is available on numerous forums, blogs and indeed in the Adwords support library.(http://adwords.google.com/support/). It is an absolute goldmine.
This is just what we found successful for this particular campaign. There are MANY other methods and testing is the most important factor in any campaign.
March 12th, 2007 at 11:44 am
Thanks for the post - I’ve been dabbling with Google Adwords for the last 18 months, but never really understood the concepts behind it until now.
Paul
March 12th, 2007 at 7:36 pm
Thanks for this post…it is very informative. Do you have any recommendations for creating instant Google Sitemaps?
-Peter
March 14th, 2007 at 6:51 am
Thank you very much for this post. I am and have been running CPC and CPM ads for various sites and markets, but after reading this I’m sure I can do much better.
Hmm, on another note, I guess Google has added a lot of information that I’ve never bothered to go back and dig through.
March 20th, 2007 at 9:50 am
Excellent post. Can’t agree enough about multi-variant testing across ALL marketing channels.
March 23rd, 2007 at 4:55 am
@Paul. You’re welcome. Let me know if you ever need any help.
@Gareth. Thank you. multi variant testing is a very powerful landing page tool.
March 24th, 2007 at 5:45 am
Despite using Adwords for a while now, there is obviously still so much to learn. Great guide for anyone new or already dabbling in PPC.
March 26th, 2007 at 8:44 pm
Thanks Leonard. It’s not really for someone dabbling. In fact, Adwords has gotten so advanced at this stage that dabbling is sometimes a waste of money. If you are going to use it yourself I think it is essential that you at least learn everything from the adwords learning center before you even open an account.
April 2nd, 2007 at 11:07 pm
I found it interesting how much improving your quality score helped lower the cpc. Thanks.
April 5th, 2007 at 6:20 pm
Thanks Jon. It may not lower your CPC, but a good quality score will certainly insure you are paying the absolute minimum for your position.
April 11th, 2007 at 4:10 am
Great post. This whole adwords QS thing has been a pain, but you laid out a great checklist to improve on it.
April 11th, 2007 at 5:09 pm
Thanks. Brad covers quality score in a lot more detail however I have just kept the basics in the form of the case study.
April 20th, 2007 at 6:19 am
Great post Dave ! Thanks for sharing all these information.
I have just one question related to this. You have mentioned about getting links with the keywords in anchor text to increase the relevance of the landing page and also adding value to their organic SEO campaign. We all know that backlinks with keywords in anchor text are good for organic SEO but if this is something they are also considering for measuring Quality score for a PPC — do we consider that they actually using a part of their organic Algo to improve their PPC relevance ?
also, a site with keywords in backlink, keywords in Meta tags and Title and keywords in content — it seems a site that is doing good in organic search is more likely to do good in PPC as well if other conversion factors are taken care of properly. Is PPC going to call for SEOs in near future ? - I would love that though :)
April 24th, 2007 at 1:15 am
Saptarshi, thanks for stopping by.
The reason I mentioned keyword rich relevant inbound links is because the whole point of quality score is to improve relevance in the sponsored listings. Google is a powerful beast and uses at least SOME form of semantic analysis to judge relevance. What better way to show Google you are relevant than to have natural relevant topical inbound links?
As for is weather PPC is going to call for SEO? Not exactly, but paid search interfaces and features are getting pretty complex. Adwords is not the “Toy” it was many moons ago. Large campaign owners are finding it increasingly difficult to keep up and are looking outside for “PPC optimizers”. (PPCOs?).
In my opinion, effective full search marketing campaign management is now becoming a little further out of the realm (Read: steeper learning curve) of the “beginner” online marketer.
By the way, I have subscribed to your blog. :)
May 12th, 2007 at 3:07 pm
Thanks for this impressive and comprehesinve post… very great so see to actual written confirmation the the organic side affects PPC quality score
be
June 29th, 2007 at 11:35 am
I love the use of graphics in this post. It’s so much more appealing. We liked it so much that we put it in our Friday “Who Said That?”
June 29th, 2007 at 5:20 pm
Thanks David. I am subscribed to your blog and have been for a while. Keep up the great work.
September 3rd, 2007 at 4:56 am
I have a new site and my AdWords ad are all getting “POOR” quality score and super high prices (1-5$) even for non-competitive terms. I’ve optimized my ads, so obviously it’s a landing page issue… but i can’t figure it out.
even for keywords that are all over my site, I’m getting a POOR rating, ads optimized perfectly and all.
do i just need to wait it out until Google re-checks my landing page? or is that a constant thing?
help!!
September 6th, 2007 at 2:09 pm
I am here in the US, Florida. Just got hired at a health insurance company and we outsource our PPC campaigns.
I will be looking after what this company do for our ppc campaigns and noticed that our landing page does not have the keywords they target on our ads.
Of course that if i would have landed this job before they selected a company to manage our ppc camp, I would have choose you guys, but now, I need to see what they are doing to either recommend another company here or to continue with them.
Anyway, my point is that we have been with this company for almost a year, and when i look at the landing page has little relevancy to the ads. Title wrong, meta tags wrong, description wrong.
I mean, we target health insurance in florida and our landing page does not have the word Florida on meta, title, or even on the body copy.
Correct me if I am wrong: don’t we need to have florida on all of the items i mentioned above? or create a landing page to target “health insurance in florida”
When I see these signs I wonder how good are they? anyone can tell me what the deal is?
you guys are my masters, you have the ONLY forum on the web that gives - let’s put it this way, i respect you because you have all the recipes to make a website successful and the proof is that your website is.
September 6th, 2007 at 2:23 pm
Hi Grace, thanks for stopping by. Enough of your flattery already! I get enough of that at karaoke!
Anyway, as I have mentioned COUNTLESS times on this blog, most of our clients have already been with another agency who completely let them down. There are WAY too many search engine marketing companies out there claiming to know their stuff and it is simply not true. On the other side of the coin, there are a lot who really do. The thing is that the best of the best don’t need to advertise and don’t need to talk the talk.
Regarding your landing pages, they should absolutely reference your keywords. If possible an individual landing page for EACH KEYWORD. Or at least each adgroup.
Without seeing the actual landing pages, I cannot tell you what they are doing wrong but if they are not tracking at the most granular level possible and have not considered quality score in their campaigns and your account is not in a constant state of flux, I would seriously reconsider. (Not pitching, I swear!). It might be worth noting, and I am sure you are aware, you are operating in one of the most competitive industries online. Your best bet in my opinion would be speaking to a super affiliate who operates or used to operate in the insurance space.
If you want to drop me a mail I will be glad to offer some free advice when I get the time.
September 6th, 2007 at 4:10 pm
Thanks! geee, i like your Irish humor, do you guys karaoke over there? :-)
Point taken, but guess what?
when i query the most competitive keywords I get to competitor’s landing pages (of very trusted or say successful ones) and when I look into their page source, almost none of them have keywords or descriptions, are they just missing your blog?…
what’s up? if health insurance is highly competitive, am I holding the gold key that nobody is catching? listen, if they give me a raise here at work because I optimized different landing pages with keywords/title/metas and body copy relating to our ads and I get a better ROI, you bet I cut you some slack and be sending you a commission! (what’s your address?):-)
your intake on this greatly appreciated
September 7th, 2007 at 5:09 am
Hey Grace,
Where did you ( I mean your company) outsource ? Though I feel sad to say, but what Dave pointed out is 100% correct - most companies that claim to be internet marketing experts often dont have the minimum clue on what they are actually doing and it is really difficult to find a good one when you are outsourcing. At the same time , the people who are actually worth their salt , in most cases, would not just beat their drums which makes it all the more difficult for prospective clients to make a good choice ( remember, not all of them are like you who reads and participates in blogs)
I think as a starting point it might be a good idea for you to ask your agency to provide month on month ROI data and ask them for an explanation if there is a low down in it or it is meagre.
Quality score is actually too important these days for any PPC campaign and you can not afford o ignore it under any circumstances. In fact, there are many companies that are employing SEOs to optimize their landing pages not because they are looking at organic rankings but more to increase the quality score of the landing pages.
September 22nd, 2007 at 5:57 pm
Hey Dave - Not sure why I had never visited your blog before, some great stuff here….
A few notes:
Adgroup size - we do it 2 ways - 1 word adgroups for any important keywords and then 15-20 keyword adgroups for lower volume or less important keywords. Taking the time to create slightly different ad copy (by putting it in its own adgroup) for a plural version of important words has paid huge dividends.
On ad creation - we have tested Sentence Case vs Title Case and almost always found that sentence case outperforms for us.
Optimizing Creative - be sure to measure conversion rate not just click through rate. As your landing page conversion rate is not set in stone, what you say in the ad can greatly move the needle good and bad. Be sure to pick the best conversion rate combined with a CTR. We measure profit per thousand impressions as a good bench mark.
Great Post and all around interesting blog.
September 22nd, 2007 at 6:26 pm
Hi Diorex, thanks for stopping by. Believe it or not but I am one of your biggest fans. Your blog (Sparse as it was/is) was one of the main inspirations for this blog. I felt bad when you took it down for that period. In fact, I was even going to approach you for a writing gig. (Not that you’d be interested.
Anyway, on your adgroup size we do something similar with the generic one worders and keep tweaking the ad text until we find a suitable CTR.
We have also tested sentence case against title case and found the opposite. I guess it depends on the industry.
Regarding optimizing creatives, a lot of times it is impossible to see at a glance which creatives are performing better. We have a tool in-house that calculates this for us. Similar to the mindvalleylabs tool. It’s based on the point on a bell graph.
Again, thanks for the comment. It is an honour.
October 16th, 2007 at 3:00 pm
Hi Dave,
Is it my imagination, or did the advert text affect the minimum bid in the past? I’ve just been looking through Google’s explanation of the Quality Score, and I’m sure it’s changed…
What does “The relevance of the keyword to its ad group” mean? Has that always been a factor?
I’ve recently been running a test on a few campaigns, changing the advert text so that the search term isn’t referenced. I’ve seen no impact on the average position or the cost-per-click, so I assume it’s not impacted my QS significantly…
October 21st, 2007 at 3:03 pm
Steve, I believe that the INITIAL quality score for each keyword is determined in part by the original ad copy and weather the keyword or synonym of the keyword is present.
October 30th, 2007 at 6:49 am
Excellent write up on Adwords quality score. I will be implementing some major changes to my campaign thanks to your article.
November 10th, 2007 at 5:58 am
I’m confused about Landing pages.
What I understand, a specific landing page should be set up for each AdGroup and they are not a part of the main website. Is that correct?
I have some ads pointing to pages on the main site and to refine the ads/destination I created special LP 1, LP2 etc.
To avoid duplicate content i have robots no follow so they don’t get crawled and muck up the SEO.
Next step is to use the split test tool…
Am I doing it right?
November 10th, 2007 at 9:32 am
Hi Roslyn, thanks for stopping by. Yes, it would appear that you are following the right steps. I believe all PPC landing pages should be kept separate from the main site and robots.txt excluded.
November 28th, 2007 at 4:54 pm
Dave, great post… I am reading it a little late but it is an excellent summary of the importance of quality score.
Somthing I have been working on for a while is a case study for advertising on branded keywords, have you or your team put anything like that together?
December 17th, 2007 at 6:15 pm
Very well written. You’ve offered a great overview on the inportance of improving your Adwords campaign Quality Score. It’s all about making the keywords relevant to landing pages. You’ve offered other great tips here as well. Good job!
I know one way that I improved CTR is by Shoemoney Adwords arrow trick. I tried it myself and it increased my CTR by 300% I’ve been showing people that on my blog and its working for others as well. Just thought I would add that tip here maybe some of you may have heard of it.
January 15th, 2008 at 4:54 pm
Excellent article. I can’t say how grateful I am for this. Thanks alot!
January 25th, 2008 at 7:37 pm
Wow, I have been looking for this kind of information for a long time.. Thanks so much!
February 1st, 2008 at 8:10 pm
How can a link building campaign help the Quality Score of my adwords campaign?
February 20th, 2008 at 2:17 pm
Dave, I like your analysis. Unfortunately, there comes a change from Google - URL displayed will have to match destination URL. Redirects are not allowed but we will still be able to use mod_rewrite.
February 26th, 2008 at 10:47 am
Hi Dave,
I’m revisiting Quality Score materials to see what other peoples’ analysis is about 9 months since things were really “shaken up”.
One interesting point you made, I’m not sure if I agree with it. Specifically, whether links are indeed a Quality Score factor.
I’ll post something up on my blog about this in the coming days.
Brent
March 7th, 2008 at 11:51 am
Hi Dave,
This is a great article & has helped me greatly. With the changes in Google QS algo, would change / add to your above recommendations?
Thanks,
Ram
March 12th, 2008 at 2:49 pm
This is really good stuff. Thanks for all of your effort in writing this post. I will be in touch shortly to let you know how well i get on. Thanks again!
March 19th, 2008 at 2:28 am
Hello Dave, when we are performing Google AdWords, we have a serous situation that most of our keywords are poor quality score.
Thanks a lot. We will implement it following your guides.
March 19th, 2008 at 6:42 pm
Hello Dave,
Thank you for sharing all of this. I have a question regarding the upper case on each word, why is this effective and should the searchwords also be written with the initial letter in upper case?
Happy easter!
Petter
March 31st, 2008 at 1:33 pm
Hello Dave,
Thank you for sharing this invaluable information.
May I ask about the range of cost for a project like this? We are a startup company, but if we could put together money, we certainly would wish to tap your expertise.
When you start a project, do you have some sense for how much you would be able to help a client?
Lastly, is there a source for CPC that people are paying for relevant keywords for quality sites?
Thank you again.
Vivek Gupta
April 5th, 2008 at 4:25 pm
Nice and useful information. I need this myself for some projects and I am always trying to figure out how to lower the CPC and higher the CTR and conversion rate. This information might help.