AdWords
1.9K members online now
1.9K members online now
Improve your AdWords performance and boost your ROI, CTR, and Quality Score
Guide Me
star_border
Reply

My quality score is rapidly dropping. Do I need to be concerned?

Follower ✭ ☆ ☆
# 1
Follower ✭ ☆ ☆

Over the last few days my quality score has been dropping very quickly. At the start the highest QS was 10 and I had many 7s and 6s and just one 3 in the keywords that I have enabled at present.

 

Now the highest QS is 5, and I have three 3s, and most of the words have a QS of 4.

 

The average position is still 1, and in competing pages it shows above all the other ads on the page. The CPC is more or less the same as it was before, yesterday at £0.42.

 

The average CTR is 14.49% yeseterday (for 68 impressions) with my best keyword achieving a 22.58% CTR (7 clicks for 31 impressions).

 

How can the QS go down so much, when the CTR is so high and I'm clearly performing better than any of the other ads for my search phrase?

It's been a puzzling time with my adwords account generally. I have been careful to select only long tail keywords with a high CTR after an experience earlier this year when the QS dropped dramatically in just one day apparently as a result of showing my ads on the high volume word "metronome".

 

I am very cautious with QS after an experience a number of years ago when my ad account QS all went so low that I could no longer run ads for any reasonable price at all. The only solution then was to close my adwords account and start a new one which immediately fixed the issue.

 

So am slightly concerned in case this is about to happen again, though I know of no reason at all why the QS should drop.

 

Before now it was increasing, it recovered from the low QS after the "metronome" experiment over a period of about a month or so. Then have just left it disabled for another month or so, while working on a new release and just enabled it again a few days ago. Was expecting the QS to stay high as before or even increase some more due to the very high CTRs my ads get.

 

Anyone got any thoughts? Do I need to be concerned, and is there anything I can do?

Incidentally, my google adwords account no longer shows the QS column. It disappeared a couple of days ago, and I can't find any way to put it back in again. It no longer appears under customize columns. So I have to find the QS by hovering the mouse over each keyword in turn now.

 

Thanks,

 

Robert

1 Expert replyverified_user

Re: My quality score is rapidly dropping. Do I need to be concerned?

Explorer ✭ ✭ ☆
# 2
Explorer ✭ ✭ ☆

First, ignore any QS when you first add keywords. I don't know if you are comparing the QS the moment you added the keywords or some time later. But the system assigns a "best guess" QS initially and adjusts if needed once it has all the information it needs, such as visiting your page to ensure you are complying with all rules. This usually happens fast - within minutes - but best to wait a few hours or even a day before telling yourself you have a certain QS.

 

>> How can the QS go down so much, when the CTR is so high and I'm clearly performing better than any of the other ads for my search phrase?

 

CTR is relative. QS is relative. That's the whole point of it. Doesn't matter if you think your CTR is high. If your CTR is 14% and your QS is 5, it means that, relative to other advertisers of that same keyword for the same position, their CTR was higher.

 

You are not performing better and that's what your QS is telling you. I don't care what your actual CTR is.

 

There's a bit more than CTR to the QS but it's mostly CTR. Maybe you have some non-CTR component where you are doing badly. I'd make sure the landing page complies with all rules. But my guess is likely CTR.

 

A handful of clicks may also not be enough for the system for your QS to be more stable. I've seen this before although it's been a long time that I've seen a QS go up and down like a yo-yo. Key here is to be patient.

 

I would not be overly concerned. You are getting traffic. I assume it's quality traffic that converts at a good ROI (good meaning acceptable to you and you're making a profit).

 

I don't know why you lost the QS column. That shouldn't be.

Re: My quality score is rapidly dropping. Do I need to be concerned?

Follower ✭ ☆ ☆
# 3
Follower ✭ ☆ ☆

Okay on your first point - yes I know about that that when first added the QS should be ignored - but these are words that have been in the account for months. I haven't added any words for at least a month.

 

Three days ago then one of them was a 10, and most were 7s and 6s with one 3. Basically over the weekend they have all gone down to best score 5, average 4 so most of them have dropped between 5 and 3 points in the QS.

 

As for relative to other advertisers - then - the thing is though - my ads go straight to the top of the search results, above all the other ads. Normally my average position is 1. Some keywords may occasionally be 1.1 or 1.5 but most are shown as position 1 so always at the top. That's on pages that may have half a dozen other ads on them, on the first page.

 

So - surely that means that my CTR must be higher than theirs? Also I know that I'm a bit biased of course, but for most musicians my ads would be much more interesting than the other ads on the pages which are often pretty generic. I tailor make the ads to the keyword. Have different ads for different keywords unless they are just minor variations on exactly the same thing. The other ads don't do that.

 

Some of the ads are for Amazon - for instance - so of course brand recognition - but are pretty boring just case of substitute keywords into a generic ad that would apply for anything you might buy on amazon.  I suspect that my ads may be outcompeting Amazon because they always appear above the Amazon ads.

 

Yes, hopefully it is just not something to worry about, and something curious about how google calculates the QS, as you say. I'm keeping a close eye on it and the CPC isn't going up at all, if anything going down, so if so then if the QS goes down and the CPC goes down maybe it's not a matter of concern.

 

But I'm slightly worried that if the QS keeps dropping to e.g. 2s and 3s, that my ads may stop running due to the QS being too low to show ads, and that at some point the CPC may suddenly go up.

 

Also at some point I'd like to experiment (cautiously) with showing my ads on the word "metronomes" (without any long tail) and so am worried there that I might not be able to get it on the first page if the QS is too low.

 

As for ROI - well it is marginal at present, but very promising. Last few days I have earned enough to pay for my ads for several more days and that's just for the immediate sales and I find for some reason for my produce, with the google search ads that many people buy after the 30 day trial period expires. Also many sales are indireclty by word of mouth they've heard of it first from someone else, who might have found it originally through a search ad.

 

So - it is a bit hard to evaluate how well it is doing. Up to now I've normally done it for a few days at a time and then stopped.

 

However, if the CPC stays as it is now, and the number of immediate sales pays back the cost of the ads and some more, then it is probably not doing too badly and if it seems as promising as it does right now, then this time, after a while I may increase my budget per day and try targetting the high volume keyword metronomes.

 

 

 

 

 

Re: My quality score is rapidly dropping. Do I need to be concerned?

Follower ✭ ☆ ☆
# 4
Follower ✭ ☆ ☆

BTW just one other thing. I left the account dormant for a month or two, and have just enabled the campaign again. The QS a few days ago is the QS it was at when I last ran the campaign.

 

So one thing I wondered about is, maybe Google has changed the way their QS algorithm works, in the last month, so that everyone's QS has gone down, and so it was just a case of it adjusting to the new norm. Or that for some reason the QS of all advertisers advertising for variants on "metronome" has gone down over the last month, and again, it was just adjusting to the new norm for advertisers of metronomes.

 

If that was it then it would be no matter for concern at all of course.

Re: My quality score is rapidly dropping. Do I need to be concerned?

Explorer ✭ ✭ ☆
# 5
Explorer ✭ ✭ ☆

>> surely that means that my CTR must be higher than theirs?

 

You appear to equate your position on your CTR. Ad position is not determined by CTR alone. It is determined by QS (mostly CTR) multiplied by your bid. So no, it doesn't mean your CTR or QS is higher. You know what your QS is (basically average or below) so you know your CTR is average or below.

 

>> for most musicians my ads would be much more interesting than the other ads on the pages which are often pretty generic.

 

Your QS disagrees. You may think your ads are better but the true measure is what those searching think by their vote of clicking on your ads. Right now, they are saying they think your ads are so-so.

 

It may not be solely your ads. You may be too broad with your keywords. If you are still bidding on "metronome", you need to remove the junk with negatives or be more specific and use keywords that describe your product more accurately, "old-style metronome" or "mechanical metronome" for example. It's not about bidding on keywords to give you high volumes, it's about bidding on the correct keywords.

 

You need to get that CTR up so the QS will go up too. This will provide many benefits, the main one being your cost going down. So think of what would make people looking for metronomes want to click your ad. What's the benefit to them to buy yours?

Re: My quality score is rapidly dropping. Do I need to be concerned?

Follower ✭ ☆ ☆
# 6
Follower ✭ ☆ ☆

Sorry, but that doesn't seem to make mathematical sense (I'm a mathematician by training). Let me explain why.

 

My highest CTR yesterday was 25%. That was for a word with a QS of 5/10.

 

It was an exact match word [download metronomes] so the negative keywords don't apply.

 

There are 5 other ads on the front page.

 

Now if my 5/10 means the others have a higher CTR than me - even if just one other advertiser has a higher CTR than me that means that in total more than 50% of all searches on that keyword ended up in ad clicks. Given that the top organic search result is "METRONOME ONLINE - free!" and many musicians how search for "download metronome" will probably look no further once they find an online metronome that does what they want to do, it is really hard to believe that 50% of them click on the ads. Especially since the other ads are all pretty uninteresting.

 

This is may ad:

 

"

Metronome to Download - For Windows - bouncemetronome.com

 
www.bouncemetronome.com/download

Any Time Signature Your Own Conductor to Help Timing!

"
Note that it has those extra links to different sections of the site - which is supposed to help with CTR. And it mentions two things that would be of special interest to musicians.
 
The other ads are:
 
  1. Metronomes at AmazonLow Prices on Metronomes.
    Free UK Delivery on Amazon Orders
  2. Download Metronome
    www.download366.com/Metronome
    Download Latest Version 100% Free!
    Download Metronome Really Fast Now
  3. Metronome
    download.cnet.com/Download
    Free Instant Download.
    2361+ Downloads!
  4. Download Free MetronomeDownload free software to
    use your PC as a metronome.
  5. Metronome Download FreeSearch multiple engines for
    metronome download free

The other ads all are unchanging if you search again. But mine rotates through several different ads all designed with great care to be of interest to musicians searching for "download metronome".

 

This is another one for instance:

 

Download Metronome - Windows Rhythms like ♫ ♩ ♪♩
 
www.bouncemetronome.com/download

Polyrhythms 4:3, 5:3 etc. Swing

Reading those ads as a musician might, mine stands out on the page as the only one that says anything interesting at all, or indeed anything at all over and above that it is advertising a metronome which you already expect.

 

I would be very surprised indeed if over 50% of the visitors to the page click on the ads (including mine). And if you understand the 5/10 as meaning that many of the other advertisers are much better than I am then presumably not just one competitor but 2 or 3 of them are all better than me. 3 of them is impossible because then you would have over 100% CTR for the ads. If two of them have higher CTR than me, then that means well over 75% CTR for the ads, which while theoretically possible, I simply don't believe to be remotely likely given the page as it is and the ads as they are.

 

And having two other advertisers at say 30% each, just slightly better than me, would mean total of 85% CTR. Two others at 35% gives total CTR of 100%.

 

So - I don't think the 5/10 QS can possibly be worked out just based on the CTR, and that the other advertisers have higher CTR than I do.

 

I've read elsewhere that some keywords have a lower QS associated with the word, due to a long run of low CTRs in the past. With those, I've read (not sure if it is true just a hypothesis) that no matter how well you do, you just can't raise them above a certain level.

 

But in the case of metronome, then I did manage to raise the QS all the way up to 10 with my last campaign run on some keywords. That was by doing just as you say very careful work on the copy, focussing on the things that would appeal and catch the eye of someone interested in my product and lead them to click through. Indeed I also had many 10s and 8s earlier in the year before my short metronome experiment caused nearly all the QSs to plummet by 2 or 3 points for a month.

 

So - and I haven't changed the ads, or added any keywords, or done anything at all to the account since the run which achieved the 10s a month or two ago. So - why if what I did then managed to raise the QS so high, why now is it so much lower, 5 points lower, with exactly the same ads on my part, the same landing page, improved product, and pretty much the same competing ads as well.

 

BTW those ads above were for the UK search results. 

 

In the US search results for the same search phrase, my ad is the only ad that appears at the top of the page on the first page for download metronomes.

 

The other ads are decidedly lack lustre:

 

Ads related to download metronomes
  1. Metronomes at Amazon
     
    Low Prices on Metronomes. Free UK Delivery on Amazon Orders
  2. Metronome
     
    download.cnet.com/Download
    Free Instant Download. 2361+ Downloads!
  3. Metronome Download Free
     
    Search multiple engines for metronome download free
     
    I simply don't believe that two of those competitors are achieving CTRs of more than 25% and that more than 75% of the visitors to that page click on the ads, but only 25% of them click on my ad.
     
    Do you see what I'm getting at?
     
    Basically the QS numbers simply don't seem to make sense mathematically, as a score out of 10 anyway. It's almost like they just "make up a number" Smiley Happy.
     
     
     
     

 

 

 

 

 

 

 

 

 

Re: My quality score is rapidly dropping. Do I need to be concerned?

Explorer ✭ ✭ ☆
# 7
Explorer ✭ ✭ ☆

You're looking at it the wrong way and thus your assumptions are wrong and why it doesn't make sense mathematically. Don't think mathematics, think standard deviations. Also think statistically significant data. Are you saying that 25% CTR is over hundreds of impressions?

 

The distribution of QS is probably close to a bell curve. That is, 60% of advertisers fall in the 4 to 6 range, another 20% below 4 and the last 20% have QS of 7 or more. Your QS of 5, even if 25% CTR long term, doesn't mean that someone with a QS of 10 has a CTR of 50%. Doesn't work that way.

 

I personally like ad 4 and maybe even 3 better than your first one. Your second is better. But I'm not the judge here (I'm not even a musician), your searchers are. I'm just one opinion vs hundreds of searchers and the majority rules. Maybe the majority like the Cnet ad. I can see how with a few tweaks you could improve your ads and possibly get a better QS.

 

Google has hinted many times that CTR is a huge part of QS. Observation by many also confirms that. I can assure you they don't just make up a number.

 

>> some keywords have a lower QS associated with the word

 

QS is based on performance. Assuming QS of 5 is the mean, even if all advertisers have roughly the same CTR, a SD calculation means that some are below and some are above. In this case, it wouldn't take much to get a good QS. Yes, I've heard of this too but I don't believe it. The math doesn't add up.

Re: My quality score is rapidly dropping. Do I need to be concerned?

Top Contributor
# 8
Top Contributor

Hi Robert, I'm going to jump in here if you don't mind as I think you're looking too closely at the math and the detail and making some fundamental assumptions about AdWords that are perhaps faulty.  Your last post is a long one so I'll try to answer it all, but I may miss some bits.

 

"My highest CTR yesterday was 25%. That was for a word with a QS of 5/10.  It was an exact match word [download metronomes] so the negative keywords don't apply.  There are 5 other ads on the front page.  Now if my 5/10 means the others have a higher CTR than me..."

 

First off, 5/10 does not mean other Ads have a higher QS than you.  In fact, regardless of your position in any particular auction, 5/10 may be the highest QS there.  Since Ad position is determined by QS and bid, it's easy to see that someone with a lower QS but a higher bid could have a higher Ad rank than someone with a higher QS and a lower bid.  

 

Before I go any further I'll also say here, probably not for the last time, that a lot of AdWords metrics are averages, not absolutes and this certainly applies to CTR, QS and bids, the three most important being discussed here.  25% would be your average CTR, 5/10 would be your average QS, etc.  Each and every auction causes a reevaluation of Quality Score so while one particular auction gives a Keyword a QS of 5/10, the next might be 6 or 4 or 3 and 9 - who knows?  It's constantly changing.  CTR also changes throughout the day and although I don't know how Google breaks down their calculations for CTR, I can be pretty confident that they don't deal in blocks as big as 24 hours.

 

"...many musicians how search for "download metronome" will probably look no further once they find an online metronome that does what they want to do, it is really hard to believe that 50% of them click on the ads."

 

Not to me it isn't!  Don't forget that a large percentage of web users don't understand AdWords and it's no disrespect to say that there's probably quite a big group of users who don't even realise some of the results on the SERPs are Ads.  Even if they do, why wouldn't they click on the Ad if it's relevant and right there under their mouse?  We (AdWords users) tend not to click on Ads much because we know it costs the advertiser money but for the average person all they're looking for is the answer to their search and if the Ad looks like the answer, they'll click it!  In fact, I could argue that if they didn't click your Ad in at least 50% of cases, your Ad copy probably isn't doing its job...

 

Although not directly related to this issue, I suspect you'll find this research interesting:  Impact Of Ranking Of Organic Search Results On The Incrementality Of Search Ads.

 

"The other ads all are unchanging if you search again. But mine rotates through several different ads all designed with great care to be of interest to musicians searching for "download metronome."


Again, here you're making the assumption that what you see is the same as what everyone else sees and is forever unchanging.  It is quite probable that if you and I performed the same search at precisely the same time on a given day that we would see entirely different Ads in different positions, even if we were both in the same geographical area.  One of the primary reasons for this is budget, another is Quality Score again.  Not all advertisers will display their Ads for each and every single possible impression on a particular Keyword.  In fact, if there are more than 11 (I think) advertisers that could trigger against a particular Keyword they can't possibly all appear on the same page since there aren't sufficient positions to accommodate that.  Even if there are fewer, budget considerations come into play as not every advertiser will have a sufficient budget to allow the display of an Ad for each and every possible impression.

 

"So - and I haven't changed the ads, or added any keywords, or done anything at all to the account since the run which achieved the 10s a month or two ago. So - why if what I did then managed to raise the QS so high, why now is it so much lower, 5 points lower, with exactly the same ads on my part, the same landing page, improved product, and pretty much the same competing ads as well."

 

Once again, I'll go back to the "things aren't static" issue and you and Lucid have already discussed or suggested some possible reasons.  I would underline the importance of timings here.  As Lucid said in one of his early replies, initial Quality Score for a newly added Keyword should really be ignored as it's based upon a best estimate by Google, not upon your own performance.  However, it's important to think about when and how this situation changes and whether it changes in a single sudden action, say when you pass a certain number of impressions on a Keyword, or whether it changes slowly over time, say a period of some weeks.  It is also entirely possible that outside influences have changed since the 10/10 scores, as you yourself suggested.  The advertising market is fluid, new advertisers may have come along, others may have left, we have no way of knowing and searching won't tell us.  Again I don't know the details but it's entirely possible the "comparitive" metrics extend over geographical boundaries and schedules, devices, etc.

 

It's also possible the QS algorithm has changed - we know it does from time to time.  

 

Is there any information to be had from hovering over the "bubble" next to one of these Keywords?  It's possible they may all simply say "Average", but if any of the options there say "Below Average" that may give a clue.  

 

Have you looked at Auction Insights?  This should really answer any questions you have about your performance in relation to other advertisers.

 

I'll leave it at that for now.  Have a look at the information in the bubble and the Insights report and then come back to us.

 

Jon

AdWords Top Contributor Google+ Profile | Partner Profile | AdWords Audits

Re: My quality score is rapidly dropping. Do I need to be concerned?

Follower ✭ ☆ ☆
# 9
Follower ✭ ☆ ☆

Hi Jon,

 

Thanks, interesting thoughts,

 

Okay let me answer a few things. First, it's not just yesterday. Every day that keyword performs very well.

 

Over all time that keyword has a CTR of 20.25%. That's for 16 clicks and 79 impressions. Also it has improved recently at any rate put a lot of work into the ads for it, and 

 

I know that is a small sample, yet, entering 79 as the sample size, and population 100000 (doesn't matter much how large) gives a confidence interval at 95% confidence of 8.83%. So - could be as low as 12% or as high as 29% but with 95% confidence is between those two and is probably more like 20%.

 

http://www.surveysystem.com/sscalc.htm

 

I agree that does give a bit of wiggle room, and if you take the lower number of 12% then myself plus one other advertiser would be just over 24%, and it would help a lot of I had more data, but then I would have to increase my budget. I could do that though just put all my budget into that one keyword for a while so it is running all the time, it is the lowest cost one at 31 cents per click.

 

Assuming it is 20% rather than 25% as our best guess so far, then - my calculation doesn't assume that there are other advertisers with 40%, so I didn't explain that too well.

 

The thing is there are - at least 4 ads on the page, including my own, on front page. Let's take the US page as I think most of my impressions are from the US.

 

To be in the middle of the curve, then surely it's not enough to have just 1 of those ads performing better than me. At least one other ad must be the same as me or better as well.

 

So me plus the other one at least equal to me makes 40% and then there is another ad better than me at more than 20%. So in total, all three ads are getting 60% of the clicks on that page.

 

BTW I keep a careful watch on the ads and continually optimise them for CTR. So the ones that have survived are the ones that have highest CTRs. So I also have a good idea of the ads that users are likely to click on.

 

I think if I just did an ad like 

Free Instant Download. 2361+ Downloads!

 

- I don't think I'd get any clicks at all, based on my experience so far. Of course I can't honestly say 2361+ downloads as I have only 1 downloadable program. But I don't see how that is likely to improve CTR to say that when the user is looking for a metronome.

 

I can of course try other variations on the ad wording and welcome any suggestions.

 

Thanks so much for telling me about the acution insights. I just tried that. Unfortunately for [download metronomes] it says not enough data.

 

But for "metronome download" - this is including use of negativve keywords of course, I get:

Display URL domainImpression shareAvg. positionOverlap ratePosition above rateTop of page rate
You96.78%1----88.46%
nch.com.au37.81%2.835.96%1.54%19.89%
download366.com33.73%3.732.30%0.00%12.74%
cnet.com29.43%3.528.41%0.78%1.09%
webcrawler.com19.33%5.418.76%0.00%0.00%
amazon.com18.47%3.318.31%1.21%4.07%
amazon.co.uk12.89%312.43%0.00%7.50%
musiciansfriend.com< 10%4.88.99%0.00%6.98%
ask.com< 10%5.18.55%0.00%0.00%
soft32.com< 10%2.88.21%0.00%2.50%


It doesn't say anything about the CTR, though. You could have those figures and still have a higher CTR for the other advertisers. Just seems unlikely. And at £1.00 a click I''m not bidding especially high.

 

Also - another point is - back to the [download metronomes] and the US page - since I bid £1.00 per click, but am charged - in case of [downlaod metronomes] only 20 pence per click, or in dollars, bid $1.53 but am charged only $0.31 per click - which is not so high a bid - and with that bid am consistently in number 1 position, surely that means that other advertisers are bidding low and not attempting to reach the top position on the page.

 

That would seem to suggest that top position - which in the US case makes it the only one at the top of the first page, rest are at the bottom of the first page - if the other advertisers have higher CTR it would seem to suggest they feel it is not worth while for them to bid higher than $0.31 per click in order to achieve top of the page position. But that seems unlikely given the increase in CTR and reduction in bids you get as a result of appearing consistently at the top of the page.

 

Do you see, at any rate it still doesn't make much sense to me.

 

Do welcome any more insights. I'd like to understand QS better. So far based on my experience I'd say it must be based on other things than CTR, and landing page. But what it is actually based on then is a mystery.

 

What I can do is to run a campaign for a while just focussing on [download metronomes] as my only keyword, put all my budget into it, for soem weeks, and by doing that could get a lot more information. It also seems likely to lead to a higher ROI. If I get no sales that way can stop it but it is worth a try.

 

I can also try a simpler generic ad making use of the word "free" a lot. E.g.

 

Free test drive, and free taster metronome, instant download.

 

or something like that and see if that or variants on it attracts more clicks than the more informative ones telling musicians about all the great features of the metronome.

 

BTW the cnet one is a bit disingenuous as their "instant downloads" nowadays bundles all the listed programs (even free open source programs, very controversially) in their ad supported download manager". There are also allegations that it secretly installs adware, e.g. here.

http://boingboing.net/2011/12/10/cnets-download-com-secretly.html

example reddit.com discussion here:

http://www.reddit.com/r/geek/comments/16971n/friendly_warning_downloadcom_adware_f_you_cnet/

 

As a developer you upload a simple installer, and they then bundle it with you don't know what.

 

I am considering just removing my programs from their database. I don't know how many people know about this, but I for one would never download a program from cnet any more. Though in the past I used it a lot. Nowadays sometimes I use it for reviews but would always look for another website to actually download the program.

 

BTW a bit of background on the ads for my metronome.

 

The features I promote in my ads - swing, polyrhythms, playing any time signature etc - are features you rarely find in any metronome, and that are of great interest to many musicians. Especially swing and being able to play any time signature (including odd). So it is no surprise to me that the ads I do that mention those features prominently are the ones that perform best in my campaigns.

 

The "your own conductor" also is very unusual for a software metronome, and likely to be intriguing to a musician, many musicians are likely to click on that ad just to find out what it means. The chances are high that most of them have never seen a software metronome with a built in "conductor" unless already familiar with Bounce metronome. There are a couple of simple iphone apps now with a related feature, with not much by way of features.

 

While - there is nothing unusual or intriguing for a musician about the other ads. Just the attraction of getting something for free. I can try an ad that promotes it in that way too but expect it to perform so badly I'll just stop running it after a while. I generally stop the ads if they get as far as 30 impressiosn with no clicks, or as far as 100 impressions with say only 1 or 2 clicks.

 

BTW is there any way to get a country break down of the impressions I wonder for adwords, anyone know? I can do it for the clicks in a rather laborious way in google analytics using the matched search enquiries - which gives the US as the top origin with 38.4% of the clicks so far in last year to today. It would be useful to help know which are the best pages to look at in the ad preview and diagnosis, and also indeed for helping to decide which countries to target.

 

Anyway thanks for your help and am interested in any other thoughts or ideas as I try to understand the mysteries of the Google QS and try to decide what is my best way of working with it.

 

BTW don't anyone say just ignore it Smiley Happy. That is what I used to do long ago and is the reason my old account got so bad that I couldn't run any ads at all and had to close it and start a new account as the only way to keep going (which immediately fixed all the QS issues and made it possible to run ads again at a reasonable price within budget).

 

Re: My quality score is rapidly dropping. Do I need to be concerned?

Follower ✭ ☆ ☆
# 10
Follower ✭ ☆ ☆

BTW I've been reading more about cnet as a result of the reply and fiollowing the links, and have decided to remove my programs from their database immediately, knew it was a bad egg, but didn't quite realise how bad it was!