Advanced Pay Per Click (PPC) Program

Certification Training
5011 Learners
View Course Now!
40 Chapters +

Working with Quality Score: Part 1 Tutorial

2.2 Introduction

Hello. This is Brad Geddes, the author of 'Advanced Google AdWords, the founder of Certified Knowledge and the PPC Faculty Chair for Market Motive. In our last video we looked at the quality score factors. What makes up a quality score? In this video, we're going to look at working with quality scores and how to improve it, and when should we improve this?

2.3 Quality Score Ranges

So, first the big question is, what is a good quality score? So, several years ago sevens were a really good number and they were fairly common. Now this has changed over time. So, right now, sixes and sevens are really good numbers and fairly common averages to see. Ten's are great to have. Tens are commonly branded terms. If you have non-branded terms that are ten's, don't try to figure out why. You'll drive yourself crazy. Smile, be happy you have tens and walk away. Now a frequency chart, eight's are actually the least common quality score. So if you don't see a lot of eight's, that's actually very common. When you get into threes and fours, you really want to work on quality score, because sometimes things like site links will stop working. Often when you get into ones and twos your ads just won't even show. So a one and a two, you must improve it. Threes, fours, and fives, it's very useful to improve. Sixes and sevens, unless you're in a very competitive industry, you can do some testing to try to increase quality score. But focus one revenue. And when you have tens it's a great indication for your keywords. Now, the one exception of this is if you have branded words which are fives and sixes. Then you do want to work on them. Because often your brand might not be being perceived well in overall search. So now when we look at our quality scores across our account one thing that you don't want to do is just look at how many keywords are in each quality score area. It's actually not that useful of data. And the reason why is because of how Google looks at quality scores to see how healthy your account is. So, what's more important is the impression number. Search engines make money on impressions. They look at their estimated cost on an impression basis. Determine what ads will get clicks. How much they'll make, so forth. So, it's not the number of keywords by quality score range that really matters. It's more about the impressions by quality score range. So when we look at this particular account, the vast, vast majority of impressions are in threes and fours. This account definitely needs some work on quality scores because they're so low. Now what often happens is when you start looking at CTRs, [INAUDIBLE] quality score ranges, you'll see a very common thing happening assuming landing pages are not issues. You'll see with twos and threes, your CTRs are really low. In this case they're under 1%. And then at fours, we have 1%. And then fives, sixes, and sevens, we're in the 5% range. Then eights, we're up to 6% range, and tens we're at 13% range. So there's often a linear progression between CTR and quality score. Now don't use these CTR ranges as saying, oh, I've got 5%, I should be a seven. Position matters a lot. So you may have a 0.5% click to rate keyword that's in position eight with a ten quality score. because quality scores are normalized by position. Usually what happens when you look into your account, you'll see a direct relationship between quality score and click duration. Assuming landing page is not a negative issue. Now, what you also want to look at is your types of terms, separately. For instance, when you look at this account, it's got roughly 60% of its quality scores at a 7 or higher quality score range. Now this as often gets into campaign organization and why you separate branded terms is for budget reasons and also easy analysis. So we're going to walk through how to make pivot tables in a separate video. But I'm going to show some pivot tables within this particular one just to give you an idea of how it works. And then we'll do all this step by step in a different one with an Excel walkthrough. Now when you make the pivot table, you can segment what data you're looking at. So in this case the data's including their branded terms for desktops and mobile devices. Now if we remove the branded terms, all of a sudden our sevens, eights, nines, and tens drop considerably in percentage of impressions and we went from roughly 60% at seven or higher. She'll run 30% at seven or higher, and now, we have almost 60% of our impressions at a four or less. So, you often do want to segment your keywords. Here's my brand keywords. Here's my non-brand. You might even segment them into, here's my brand keywords. Here's my product plus other people's brand keywords. Here's just product based words or service based words. Maybe here is long tail. We need to do the analysis to see if your percentage of impressions from score ranges varies dramatically by type of keyword. But, from an overall account standpoint, this is what Google looks at. What's your percentage of impressions by quality score range and that'll give you a good idea of how good or how poor your account is on quality score purpose.

2.4 Diagnosing Where to Start

The one thing you can't do within AdWords is see historical quality score information. Whenever you run a report in Google, the quality score displayed is the quality score of a keyword at the moment you ran the report regardless of the time frame you're running the reports in. So what's useful then is to schedule keyword reports on a frequent basis, such as once a week and sent to you, then you can download these so you can archive your quality scores by date range. The reason is important when you note that's a keyword has dropped in quality score. Now you have an isolated time frame to see if something you did cause that particular issue. So one good tool for this is the Change History Tool. You can go on to the tool, you can set a time frame then you can see changes within that time frame. So often what effects quality score are changes to the ads themselves. It could also be a change to a landing page. So by looking at changes in the time frame it dropped, and you maybe able to know why it changed. This is also while using the AdWorks Editor. They have a notes feature. It's useful to keep notes. About changes you made to see if that did cause a cause core change. And we're really organized and you can keep, history notes in a project management system or a team shared calendar so that you do have changes by time frame. And if you see cause cores drop or even conversion rates drop, any metrics change you have historical data to be able to isolate a timeframe change and then see what changes occurred in that particular timeframe. So when we're trying to increase quality scores, what we don't want to do Is go through every single keyword, look at every single keyword's quality score information to find areas of improvement. That can be useful for some of your very top terms but if you have a larger account trying to find those individual spots is really difficult. So what can be more useful is to take your data and put it into a pivot table and again, we'll walk through pivot tables in a different video, and then look at your average quality score by Ad Group. And how much money that Ad Group has spent. Now I'm looking at just some cost how much an ad group has spent and your average quality score then you can find ad groups that have low quality score and highest amount spent. And then you can go to the ad groups themselves to start seeing which keywords do have those low quality scores and walk this optimization we're going to get to in a moment. Now this is optimizing based upon current spec. The issue is there are times that you'll have keywords with such low quality scores. They'll never get displayed, so that can't spend money. So then there's a second analysis you can do. We can again use a pivot table. Look at how many keywords have low quality scores by Ad Group. [00:03:04]So you can go in a pivot table, you can then filter and just show me keywords that are 1, 2, and 3 quality scores. And then we'll have a list of our ad groups and how many keywords are in that low quality score range by Ad Group. So these two analysis types then, looking at spend by ad group, with low quality scores, and looking at ad groups with multiple keywords with very low quality scores, give you a starting point. And then you can go to those particular ad groups to work on quality scores.

2.5 Increasing Quality Scores

Now, the major issue with most low-quality score accounts has to do with keyword organization. And so, what commonly happens is, someone has an ad like John's Plumbing Service, serving Chicago area, and their keywords are Chicago plumbing, plumbing services, broken pipes, emergency plumbing, and like, well, I'm a plumber so, of course, I do all of these items. Now, from a searcher prospective, if you do a search for emergency plumbing, and you see two ads, and one says, hey, I'm John's Plumbing, the other one says, no, I'm an emergency plumber. That's a really big difference, and the one with emergency plumbing will often get a better click-through rate and better quality scores. So, Stop-Wanted is asking yourself, have we done good granular organization between our ads and our keywords? If we haven't, that really is step one. So, if you look at your low quality score keywords, ask yourself, could I have a better ad for this keyword? If yes, then you want to move that keyword or those keywords together to a new ad group and write better ads for those particular keywords. Now when you're moving keywords around for quality score purposes, just move the low keywords, so in this ad group, we have a bunch of twos and a bunch of sevens. We're going to leave the sevens where they are, they're an okay shape from a quality score standpoint. We're going to move the twos to a new ad group, and then rewrite ads. So when you find low quality scores, assuming it's not landing page, and we'll touch landing page in a few minutes, what it really comes down to is ad testing for CTR.

2.6 Testing Ideas: Ad Type and Serve Rate

Now when we look at what CTR affects, we have two different factors here. We have one which is expected click-through rate. This just means your overall CTR is lower than expected and testing ads to increase your CTR, will increase your quality scores, in most cases. The other one we have is ad relevance. Now, ad relevance is how related your keywords in your ad are to each other. Now, these are still a set of CTR-like formulas that are sort of attempting to automate the user experience from a relevancy standpoint. Just raising CTR may not increase ad relevance because some aspects of the ad, Google does not consider related. So what you really get into a testing then is with expected CTR below average, you could do incremental testing. You could start with your current ads, rewrite just a description line, change some words around, and then test that new ad. With ad relevancy below average, you often want to start with a brand new ad or radically change your ad around because your current ad could have a really high CTR. But if its ad relevance is below average, there's aspects of the ad Google doesn't consider relevant. So you want to start with a very different ad. Now, even with expected CTR below average, you can start with complete different ads too, you don't have to do incremental testing. With expected CTR, you're just doing ad testing, whether they're slight changes or major changes to try to increase CTR to increase quality scores. With ad relevance below average, then you really want to try completely different ads. Now one thing to look at though is your quality score is a roundup of all the ads that keyword is being displayed for. So what might happen is if you go into an ad group and you have a keyword with a low quality score but you have two or three or four ads in that ad group, what you're really seeing is the average of all the ads combined to that keyword's quality score. So then, look back to the ads that are currently running to see if there's major CTR differences in the ads. So what Ad Serve Percentage is, is a good proxy to use here. It's the percentage of times that an ad was served from that ad group. So in this case, we have one ad with a 14% click-through rate, it's only being served 14% of the time. We have another ad, which is a 66% ad serving percentage, it served the majority of the time, with a lower click-through rate, it's 11.44. So if we just deleted our two lower CTR ads, we would see our CTR go up for this ad group, without actually changing anything else in the keyword, or moving the lower CTR ads. So this gets into your ad serving type. So in this particular account, they are optimizing for conversions. So the ad that's being served at 66% of the time, the reason why, is it's the best conversion rate, it's not the best CTR. So you do have a balance between conversion rates and quality score in some cases, and we're really going to get into that in a little bit later. But if you were optimizing for clicks, and you see that you've got lots of ads or maybe you haven't ended ad tasks, and so you have a lot of low CTR ads that are being served that maybe aren't your best conversion rate ads either. Just by deleting those you can increase the CTR for that keyword without actually writing new ads.

2.7 Testing Ideas: Device Type and Relevance

Now when we get into ideas for writing ads, we have an entire video on ad testing ideas from copy standpoints. But something to look at for quality or specific is, if your campaign is being shown on all the device types. Computers and mobile devices, but do you have a big CTR difference by the device type? Because again quality score again is technically different by device. So we look at these two particular devices, same keyword, same even average position, actually between devices, are mobile ad is almost a 6% CTR. Our desktop add is a 1.95% CTR. So in this case, it's not necessarily just about changing the ad around. What we'd want to do, is make sure that we have a mobile preferred ad. That's the ad that's getting 6% CTR. And if we haven't set mobile preference, we'd set mobile preference on that ad that's got the high CTR for mobile devices. Then we would create another ad for the desktops that is not mobile preferred. So, if you've already done at two or three mobile preferred and two or three desktop ads, if you have major CTR differences between the device types you're just going to work on the ads for that particular device type. If your CTR's are similar by device type, then go ahead and change all the ads around to do your testing. Now you're going to see where average position matters a lot because if your average position on a phone is four you're usually at the bottom of a page. So your position is four on a computer, you're still above the fold. So in the case like that, your desktop CTR may be dramatically higher than your mobile CTR, but it's so your desktop ad that's the issue, not your mobile, because your mobile is bottom of the page, of course it has a lot lower CTR. So you have to take some position into account here. But if you do see dramatically different CTRs by devices then you may want to adjust to your ad testing by device. Now also, examine your current ads. In this particular Ad group, we have three ads. One has a 13.89% CTR. The next one has a 1.16% CTR and a 2.35. So we have one ad here that is dramatically better than the others. Its average CBC in fact, is much lower than the others. For these ads again, the position is very similar. So what's happening is the top ad with the 13 3% CTR, has a much higher quotas or combination with that particular keyword. So in this case, just the leading are two lower CTR ads would again increase our CTR. I might fix our quotas for issues. But, if we saw that that didn't work, or we had some keywords that were just not relevant to our ads, we could take those keywords, move them to new ad group, take our ideas and our really high CTR ads for our ad testing. So don't forget to learn from your account when you see ads with much higher CTRs, to learn from them with ad testing. So, when you get into items such as ad relevancy and expected CTR, step one is asking yourself could I have a better add for this particular keyword? If the answer is yes and you have other keywords in an ad group that don't have any relevance to the ads, move the low performing ones to new ad group and do ad testing. If the answer is no, you're like these ads are pretty good, then you're just going to do ad testing within that exact same ad group to try to raise CTRs. Just make sure you look at other ads to get ideas, and you look at your device percentage click-through rates as well, and any ad serving issues that might be occurring.

Find our Advanced Pay Per Click (PPC) Program Online Classroom training classes in top cities:

Name Date Place
Advanced Pay Per Click (PPC) Program 5 Jun -20 Jun 2021, Weekend batch Your City View Details
  • Disclaimer
  • PMP, PMI, PMBOK, CAPM, PgMP, PfMP, ACP, PBA, RMP, SP, and OPM3 are registered marks of the Project Management Institute, Inc.

Request more information

For individuals
For business
Phone Number*
Your Message (Optional)
We are looking into your query.
Our consultants will get in touch with you soon.

A Simplilearn representative will get back to you in one business day.

First Name*
Last Name*
Work Email*
Phone Number*
Job Title*