homepage Welcome to WebmasterWorld Guest from 54.205.205.47
register, free tools, login, search, subscribe, help, library, announcements, recent posts, open posts,
Subscribe to WebmasterWorld
Visit PubCon.com
Home / Forums Index / Google / Google AdWords
Forum Library, Charter, Moderators: buckworks & eWhisper & skibum

Google AdWords Forum

    
Quality Score vs Quality of Traffic
briggidere




msg:4178376
 5:28 am on Jul 29, 2010 (gmt 0)

This is one thing that has intrigued me for a while now.

There are many methods that Google use to track us around the web, cookies, toolbar, analytics, signing in etc.

Could there be something in the algo that tweaks the results based on the users previous patterns. eg: past successful conversions.

Lets say we have 2 advertisers bidding on the same term.

Advertiser A has a 8/10 QS
Advertiser B has a 4/10 QS

Both get the same CTR, have similar average positions & Conversion rates.

"This is where I got my brain working in overdrive"

User X has a history of many past successful conversions (as recorded by Google)
User Y has never recorded a conversion, but does as much online as user X so Google has data, just no conversions.

Let the searching begin:

User X does a search and Advertiser A's ad is displayed as normal, but not Advertiser B.

User Y does the same search, but this time Advertiser B's ad is displayed with Advertiser A's ad not showing.

This could mean that the higher the quality score of your keywords, the more likely it is you will receive a higher quality visitor from AdWords. I am not talking about bid prices, just the QS vs traffic quality

I know this is all speculation but has anyone seen a correlation between QS of your campaigns and the quality of traffic it brings in?

 

dougmcc1




msg:4178577
 2:37 pm on Jul 29, 2010 (gmt 0)

No

netmeg




msg:4178612
 3:21 pm on Jul 29, 2010 (gmt 0)

Domain parking relies on traffic quality. I would not be surprised if it were an element in both AdWords and AdSense.

LucidSW




msg:4178803
 7:57 pm on Jul 29, 2010 (gmt 0)

> Advertiser A has a 8/10 QS
> Advertiser B has a 4/10 QS
> Both get the same CTR
> similar average positions

By definition, this scenario could not possibly happen. If they have the same CTR at the same position, their QS would be similar.

Being in the same position, B would have to bid twice as much. However, the QS is saying they wouldn't get the same CTR. B's ad is of poorer quality and thus the click, while they can be valuable, is not as valuable as A's, not of higher quality. A is getting twice the number of clicks for the same number of impressions (at about the same cost). At the same conversion rate, he's selling twice as much so his ROI is double.

Your QS is your indication of quality. It's called _Quality_ Score for a reason. Basically, the masses have voted with their clicks. The system works and there's no way you'd somehow get better quality clicks. Only if your conversion rate was at least twice that of A's, which is highly unlikely, could B feel good about his campaign. But then, if he increased his QS, he'd do much better still.

The answer to your question, which you already answered is, the higher quality traffic comes from your higher QS keywords. There is no question about that.

ppc_newbie




msg:4179006
 5:54 am on Jul 30, 2010 (gmt 0)

The answer to your question, which you already answered is, the higher quality traffic comes from your higher QS keywords. There is no question about that.

Except for G not having all the typos/mispellings/etc. figured out quite right for QS yet.
ie. Nice Ad/LandingPage for "red widgets"
QS = 8 for keyword "red widgets"
QS = 3 for keyword "red widjets"
even if there is good volume for "red widjets" it may kill my ROI to up bids

exmoorbeast




msg:4179072
 10:04 am on Jul 30, 2010 (gmt 0)

Absolutely agree with PPC newbie. Google hasn't figured out misspellings and there are 00s of other QS anomalies that we have found.

Besides someone can have a low QS and be next to a person with a high QS, they just have to bid a whole load more. You can be QS 3 and be number 1. Let's not forget that!

LucidSW




msg:4179207
 2:35 pm on Jul 30, 2010 (gmt 0)

You guys don't get it, just like most advertisers.

>> Except for G not having all the typos/mispellings/etc. figured out quite right for QS yet.

It's not Google that sets the QS, it's the marketplace. If your misspelled keyword is getting a poorer QS, it's because searchers don't click your ad. Maybe you should try a different ad for your misspellings so they get clicked on more. Remember, QS represents how good your CTR is compared to competitors.

Yes, you can be in first place with a low QS by bidding high enough. Ranking is calculated by multiplying bid and QS. So with a QS of 3, you'd have to bid more than 3 times what a QS of 10 is bidding. Of course, expect to pay close to your bid as well.

mrguy




msg:4179294
 5:06 pm on Jul 30, 2010 (gmt 0)

It's not Google that sets the QS, it's the marketplace. If your misspelled keyword is getting a poorer QS, it's because searchers don't click your ad. Maybe you should try a different ad for your misspellings so they get clicked on more. Remember, QS represents how good your CTR is compared to competitors.

Yes, you can be in first place with a low QS by bidding high enough. Ranking is calculated by multiplying bid and QS. So with a QS of 3, you'd have to bid more than 3 times what a QS of 10 is bidding. Of course, expect to pay close to your bid as well.


That is how it's supposed to work but I've got numerous anomalies where I have a very high CTR, but the QS is just so, so.

The problem is the algo is not perfect and it does screw up more often than people think.

LucidSW




msg:4179313
 5:42 pm on Jul 30, 2010 (gmt 0)

Define "high" CTR. You can't compare absolute CTRs with each other.

In a niche where the CTR in first position averages 2%, a 3% CTR will be considered high and your QS will reflect that. Conversely, in a niche where a 12% CTR in that position is average, your 10% CTR which you find high, will be calculated to an average or low QS. It's all relative and that's how QS is calculated.

exmoorbeast




msg:4179344
 6:07 pm on Jul 30, 2010 (gmt 0)

Lucid, are you aying the there is any different intent to click an ad for [lonas] as there is [loans]? Sometimes advertisers are too smart for the system and that's something that is often discussed at conferences. This could be a good example.

As to us 'not getting it' I've spent more money on Google than you probably ever will, and have been very successful at that!

mrguy




msg:4179355
 6:19 pm on Jul 30, 2010 (gmt 0)

In a niche where the CTR in first position averages 2%, a 3% CTR will be considered high and your QS will reflect that. Conversely, in a niche where a 12% CTR in that position is average, your 10% CTR which you find high, will be calculated to an average or low QS. It's all relative and that's how QS is calculated.


Forgive me, I forgot Google Adwords is perfect and never ever makes a mistake much like their organic search is perfect and never makes mistakes.

I've been doing this long enough to know what a high CTR is for the sites that are being promoted is.

Sorry I brought it up.

LucidSW




msg:4179464
 9:44 pm on Jul 30, 2010 (gmt 0)

> Lucid, are you aying the there is any different intent to click an ad for [lonas] as there is [loans]?

There may very well be. I make typoes sometimes. Google comes back with "did you mean 'loans'?" I click the link and get new results. That means I did not click on the ads thus contributing to lowering your QS. It's not that I did not intend to click ads for [lonas]. I simply realized my mistake and corrected it before I had a chance to click on ads.

Channel01




msg:4179650
 8:32 am on Jul 31, 2010 (gmt 0)

The answer to your question, which you already answered is, the higher quality traffic comes from your higher QS keywords. There is no question about that.


I think Google has pioneered a great system but your view is a little utopian, imo. My experience is that it’s dangerous to assume that Google’s definition of quality is directly in line with the advertiser’s definition of quality (traffic). I can think of a few cases where it’s not always true.

For instance, say a company sells widgets but they tend to be of the premium variety. The advertiser may very well want to be in the auctions for generic terms (to capture volume) but they don’t want traffic that is price sensitive. If the company really wants to ensure quality traffic then they’ll use premium messaging in their ad copy to pre-qualify the traffic. The problem is that this will likely discourage certain traffic segments from clicking on their ads. This will, in turn, reduce the advertiser’s CTR and thus their quality score. The traffic will be highly relevant to what the company offers though. So this an example where a better QS from Google doesn’t always mean that the advertiser will receive a higher quality of traffic.

As another example, let’s say a company improves their QS for a term which allows them to achieve a higher rank for a similar max bid. It’s very possible that the improved rank could enable their ad to enter into more auctions with Google’s partner network (since many partner sites only display the top x ads). Now, I’m not sure what the experience of others is here but my own experience is that, on average, the partner sites tend to send a poorer quality of traffic than Google refers from its domain. So, it is possible that improving QS/rank/traffic can actually worsen the quality of an advertiser’s traffic if it leads to a spike in partner network traffic.

Overall, these are just two examples but I just think it’s too simple to imply that a higher QS = higher quality traffic. What a strong QS really tells us is that users (and thus Google) think the advertiser’s ads are highly relevant to their search queries. Whether or not the user actually finds the advertiser’s website to be relevant to their searches is an entirely separate question. It’s an important one though because it ultimately determines the quality level of the traffic the advertiser is receiving.

Soccout




msg:4179652
 8:39 am on Jul 31, 2010 (gmt 0)

Though I think, that this definetely is not true, i also thought that the QS is not just CTR correlated with the position. Its also the Quality of your site, isn't it?

Swanson




msg:4179783
 2:53 pm on Jul 31, 2010 (gmt 0)

Yes, LucidSW is missing one fundamental part of Quality Score.

It is not about the marketplace and CTR per se although that is one component - it is scored by a landing page bot (Adbot) that crawls the landing page and looks for the relevance to the bidded keyword and the quality of the content.

So it is not true to focus on the "marketplace" and CTR of the ad in relation to quality score - anything based on that purely is bad advice.

Swanson




msg:4179788
 3:06 pm on Jul 31, 2010 (gmt 0)

I just wanted to ad that the quality score is not simply what users think of your ad - it is really mainly an algorithm and calculated by Google, as I said before the marketplace does not really decide the quality score.

I have pasted the info from Google's help pages on how is quality score calculated:

•The historical clickthrough rate (CTR) of the keyword and the matched ad on Google; note that CTR on the Google Network only ever impacts Quality Score on the Google Network -- not on Google
•Your account history, which is measured by the CTR of all the ads and keywords in your account
•The historical CTR of the display URLs in the ad group
•The quality of your landing page
•The relevance of the keyword to the ads in its ad group
•The relevance of the keyword and the matched ad to the search query
•Your account's performance in the geographical region where the ad will be shown
•Other relevance factors

As you see only the first 3 mention CTR (i.e. the user) - and one of those is based on your account CTR (i.e. how good are you in general in managing your account and ads)

LucidSW




msg:4179794
 3:49 pm on Jul 31, 2010 (gmt 0)

> The advertiser may very well want to be in the auctions for generic terms (to capture volume)

That's just it. Don't advertise to capture volume. Do it to capture QUALITY traffic. I've tried to explain this to clients many times. Their thinking is, the more the better and we'll sell more. Well, they will sell more but not proportionately more. The higher returns are always on the quality keywords. If you sell premium widgets, don't bid on generic widget keywords. Bid instead on "premium widget". Sure, you won't get nearly the amount of traffic. But that traffic will be of higher quality to what you sell thus resulting in better ROI. The more generic terms will get some traffic, maybe even a lot, and some will result in sales but the overall ROI will be reduced.

If you bid on more generic terms and have an ad that mentions a premium product, less people will generally click. That is what you want in your example, trying to get the more qualified traffic with your ad. But you also have to do it with your keywords. That's why Google shows QS at the keyword level unlike Yahoo which shows it at the ad level.

As for improving QS getting you better position and showing more on the partner network, I agree. It usually is of less quality. If that's the case, turn off the partner network. Of course, that's at the expense of less traffic and some of those would convert. But overall, your ROI will improve. Isn't that the bottom line, not the actual number of sales?

As for quality of the site, that's a separate issue. Adwords is there to get traffic to your site. Once you get it, they've done their job (your job really since you are in control). If that traffic doesn't convert because you are not bidding on proper keywords, it's not Google's fault. If you do get quality traffic but it doesn't convert, again, not their fault.

I know that a significant portion of QS is keyword relevancy. But there is no reason not to bid on irrelevant keywords. None. What's left is mostly CTR. That's what QS is all about. So if you want a high QS, work on your click rate for your relevant keywords. The marketplace tells you if they like your ad or not. It is a vote by the users.

I'm really surprise to hear Swanson say that the marketplace does not decide. Everything he posted from Google's help point to that. That includes keyword relevancy because if your ad shows for an irrelevant keyword, users just won't click. The quality of your landing page is simply if you are following policies. As for account history, I've never understood that part and really don't believe it has a big impact.

Swanson




msg:4179810
 4:34 pm on Jul 31, 2010 (gmt 0)

LucidSW, Google chooses the relevancy of the keyword to the landing page and dynamically to the query term.

Google uses a semantic algorithm using technology to do this. The marketplace does not decide in this portion of the algorithm.

Edwin




msg:4180047
 5:19 am on Aug 1, 2010 (gmt 0)

The more generic terms will get some traffic, maybe even a lot, and some will result in sales but the overall ROI will be reduced


Surely if the ROI is still positive even for the reduced ROI search terms, the savvy merchant will want those sales too? It's only if the ROI turns negative that they should stop bidding and sharpen their focus. Profit's still profit, even if it's reduced profit!

LucidSW




msg:4180131
 1:14 pm on Aug 1, 2010 (gmt 0)

Those generic terms may have a negative ROI. You don't want those, do you? You want to maximize your profits. Therefore, you remove anything that negatively impacts the bottom line.

netmeg




msg:4180133
 1:33 pm on Aug 1, 2010 (gmt 0)

People either get it or they don't. It's as sinple as that.

buckworks




msg:4180208
 4:54 pm on Aug 1, 2010 (gmt 0)

Regarding those broad, generic keywords that you're bidding on just to get your name out there, I find it can work well to put those into separate campaigns from the premium searches with the stronger ROI. That gives you finer control over the bidding.

It can be useful to run ads for brand exposure as long as you find the bidding vs. conversions balance that keeps those campaigns above the break-even point.

Channel01




msg:4180859
 12:00 am on Aug 3, 2010 (gmt 0)

Those generic terms may have a negative ROI. You don't want those, do you? You want to maximize your profits. Therefore, you remove anything that negatively impacts the bottom line.


Not necessarily. I've had clients similar to my description above where generic terms do quite well for them from an ROI perspective. This is especially the case when we're talking about a high end business where the company can offset lower conversion rates via high average values.

ppc_newbie




msg:4181582
 3:28 am on Aug 4, 2010 (gmt 0)

But if my positive ROI on generic KWs is still acceptable, I will take the extra volume.

More money NET is still more money in my pocket.

LucidSW




msg:4182018
 7:08 pm on Aug 4, 2010 (gmt 0)

> More money NET is still more money in my pocket.

Really? Depends on how much you pay per click for the traffic and the conversion rate for each keyword. Let's take two examples. I created the following scenario (not real world but could be typical).

keyword #1 (very relevant) results in 100 clicks/1000 impressions (10% CTR), costs $0.30 per click and the conversion rate is 8% (8 sales total).

keyword #2 (NOT relevant) results in 100 clicks/10,000 impressions (1% CTR), costs $0.60 per click and the conversion rate is 1% (1 sale).

Using both keywords gives 9 sales. The total Adwords expenditure is $90. Let's say the products sells for $49. That's $441 in revenues less $90 = $351 net profit.

Using only the very targeted keyword is 8 sales. Total Adwords expenses are $30. You make $392 in revenues less the $30 = $362.

You make $11 more net by using only the targeted keyword, even though you make more sales by using both. If I have a keyword that is less relevant, getting a _much_ lower click rate, costing _more_ and with a _much lower_ conversion rate than targeted keywords, I'm only hurting my bottom line.

I know, we're talking about generic keywords here, not irrelevant ones. But the same principle can still apply. If you sell books and bid on the kw "books", I doubt you'll have a very high conversion rate. It will also cost you more per click than the more specific "sci-fi book" and the even more specific "isaac asimov books" will do much better still.

buckworks




msg:4182029
 7:16 pm on Aug 4, 2010 (gmt 0)

That's $441 in revenues less $90 = $351 net profit.


You can't call it net profit until you account for the cost of goods sold.

If a generic or less relevant keyword converts poorly, you should be bidding less for those clicks, not more.

It's essential to know your break-even points for different campaigns, so you don't pay more for some traffic than it's actually worth to you.

LucidSW




msg:4182533
 3:30 pm on Aug 5, 2010 (gmt 0)

Of course, that should say gross, not net.

If a generic or less relevant keyword converts poorly, that should tell you that maybe that keyword should not be used. Or bid less as you say. The effect of that is a lower position, likely off the first page thus resulting in less impressions and less clicks, going against the original (Channel01's) plan of trying to get more traffic. If someone clicks your ad, they likely already have clicked others and may already have made their buying decision. My data shows lower conversion rates in lower positions, about 10 times less on the second page and a CPS about 6 times higher.

Also, your QS is likely to fall thus meaning you'd have to increase bids just to maintain your position or fall further back.

Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Google / Google AdWords
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About
© Webmaster World 1996-2014 all rights reserved