Poor Keyword Score

quality,score,keyword,adwords,poor,ctr

[i9seno]

I got this interesting issue with quality score.

An add based on a single, very targeted keyword. Quality score of 4. Poor keyword relevance.

Now, the keyword issue is highly bothering me. I suppose it comes from the low CTR, in terms, lowering the quality score.

So, we are talking about a very specific keyword, rarely used in search querries, one that mostly people who are really interested in the service I offer would type.

My first idea was - change the add slogan, in order to urge people to act(click on it). This should have increased the CTR; thus the quality score. This should have eliminated the issue with poor keyword relevance.

It didn`t...

So, can someone suggest something else(besides modifying the add text itself) that I can do to get my add back in shape with higher quality score.

Thanks in advance!

p.s.
A few weeks ago - the CTR was much higher, the quality score was 2-3 levels up. There was no "poor keyword relevance" issue.

[buckworks]

You should definitely test some variations in the ad text to see if that improves the CTR.

Also, check your query reports and watch for queries to block where your ad is off-target even if there's some vocabulary overlap. Tightening up your targeting is another way to improve performance.

Make your decisions based on cost per conversion, not just cost per click. Occasionally it's possible to come up with an ad that converts better even though AdWords doesn't assign it the best quality score.

[LucidSW]

I don't think it used to be the case last year but now, if your CTR is below a certain threshold, the system will change your keyword relevancy to poor. At least, that's what I understand talking to some Adwords reps and reading between the lines. It's not just the relevancy to your page.

I've had the same thing happen except in reverse. Definitely proper keywords for the product (service actually) and given poor keyword relevancy. A local service and few competitors, almost always in first position. CTR was not that high for being first, in the 2% range. Then, with another ad, it shot up to 10% and the QS went from 3 to 7 and the keyword relevancy is now good which is totally against what I used to think, that relevancy cannot change.

Your ad is what determines CTR. You may think it's good and that people will click on it but the proof is in the pudding. If they don't click, it's not good enough. Try other ads.

You really should base decisions on CTR x conversion rate. It will be rare that a poor CTR (and thus poor QS) ad gets a high conversion rate. I can't remember it happening to me ever. Higher CTR and higher conversion rates seem to follow each other so you'd be leaving money on the table. I'd much prefer an ad with 10% CTR and 5% conversion than one at 1% and 20%.

[GetReal]

I like to call it the 'Google Washer Machine...'

[i9seno]

Hi, thanks for all the responses.

It did work for one campaign, but for the other...

Will give you an example of what really leaves me clueless:

A website for ordering drinks.
Two pages for adwords - one for fanta, one for sprite.
Both pages are 100% identical, except for one including fanta on several places, the other - sprite.

Adwords groups - one for fanta, one for sprite.
Keywords for groups -1.fanta;2.sprite.
100% identical ads - one with Drink Fanta, one with Drink Sprite.

So, the fanta keyword has a quality score 4.
The sprite - 6.

How is this possible when everything else is virtually the same?


p.s. I`m not running the above campaign, but what I`m doing really shows up the same results as above. Can anyone give me a clue of what`s happening?

[lucy24]

How is this possible when everything else is virtually the same?

How closely analogous is your real situation? In this example, the obvious difference is that the names themselves aren't equivalent. "Fanta" in English is only a brand name; "sprite" is also a word with other meanings. So it's just the kind of thing that a brainless computer would get hung up on.

[LucidSW]

>> How is this possible when everything else is virtually the same?

That's just it. They are not the same.

You need to understand how QS is figured out. To make a long story short, your CTR is compared to that of the historical average of every one else using that keyword. In mathematical terms, a standard deviation (SD) calculation is made against the average and your own CTR determines how far away from the mean you are. So if the mean CTR is 3% representing a QS of 5 and the SD is calculated as 0.5, if your CTR is 4%, you are 2 SDs from the mean or a QS of 7. If it was 2%, you are still 2 SDs away but going the other way or a QS of 3.

Since the historical CTR of all those using the Sprite keyword is different from those using the Fanta keyword, the mean QS is starting from a different base level. The SD calculation will also be different so of course your QS will be different, even if they have the same CTR.