Forum Moderators: buckworks & skibum

Message Too Old, No Replies

The Quality Score in Action.

         

luke175

9:25 pm on Nov 14, 2006 (gmt 0)

10+ Year Member



So all my bids are at $5-$10. My site is PR5 and each ad group is relevant and I only use 3-8 keywords in each adgroup. No adsense or affiliate links.

So I'm totally shut out.

So who's in first place now for my biggest keywords? One of my affiliates selling my product. So apparently an affiliate is higher quality than the actual merchant.

Spot 2 is a total MFA site and spot 3 is "Buy KEYWORD on ebay"

Quality!

rbacal

6:11 pm on Nov 16, 2006 (gmt 0)



Taking that advice - you'll have ignored almost all of the greatest successes in history. Almost any successful person you talk to will say that they've had some major failure moments that they have learned a great deal from.

Of course there's failure moments, but I want to talk to people that succeeded by overcoming those failures. I don't know about you but if I want to "kick alcohol", I really don't go to the drunks on the street to learn how to do so.

If they can only talk about how they failed, and fallen and couldn't get up, and haven't gotten up, all I'm going to learn from them is how to be like them.

No matter. I understand that there's a lot of people who would rather commiserate with other people who are in the same sunk boat, then actually try to figure out why there are "those other people" merrily floating down the river with big smiles on their faces.

Philosopher

6:23 pm on Nov 16, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Actually, you jumped into this thread and basically began talking down to the other posters. That doesn't really help those that are trying to figure things out.

Instead of just talking down the the posters, why not offer some advice? From your posts and tone you seem to understand exactly what is going on, so please enlighten those that aren't so fortunate.

Also if you would read the OP's first post and his/her subsequent posts, you might realize something that you seem to have missed.

The poster is trying to figure out why his keywords pointing to page X have been hit with the $5 and $10 CPCs, when one of his affiliates [bold]who is linking to the OP's own site/bold] is now the top advertiser on those same keywords.

If the OP's QS is low because he has a "sales page", then the affiliates should be as well. It's not, which says a lot about the QS.

From the OP's original post, it is a little unclear whether the affiliate is linking to the exact same page that the OP was, or just to the OP's site. I would like clarification on that (although I would imagine it's the same page).

pdivi

6:34 pm on Nov 16, 2006 (gmt 0)

10+ Year Member



I don't know about you but if I want to "kick alcohol", I really don't go to the drunks on the street to learn how to do so.

Nor do you go to people who have never tried alcohol. I think the first step here is to look at the outcome of the QS in a factual way, then make tracks back to how it might be arriving at that result. The fact is, all we have on how the QS works is some very vague guidance from G. So to say the QS drives off of things like privacy policies and domain registrations using a complex multivariate algo to optimize user experience is as much a guess as me saying it drives off of a very simple bid behavior algo to optimize short-term revenue. So, it's wrong to say that we're wasting our time by discussing theories and observantions in lieu of acknowledging The Truth, as defined by the those who were not hit. Observations are the only way we're going to find out how to profit from this.

I've spent a lot of time talking to people I know in the industry, and I'm reporting back my findings. I'm weighing my findings against reports from people like Green_Grass, who got hit and managed to make it back. But the answer here is in aggregating all of the observations, not just those of people who have been smiling all the way down the river.

For all I know, they're just smiling because they don't know how to optimize their CPA properly.

venrooy

6:45 pm on Nov 16, 2006 (gmt 0)

10+ Year Member



I don't know about you but if I want to "kick alcohol", I really don't go to the drunks on the street to learn how to do so.

But if I wanted to prevent alcoholism - I'd go to the drunk to ask him how he got there in the first place. There is something to learn from everyone.

You're falsely assuming that everyone that complains in this forum are complete Google - and Business - Failures. You couldn't be further from the truth. Most of us air our complaints here, so that others can help us see them from different angles, and so that others are prevented from doing what we have done.

Assuming that we are failures, only exposes your ignorance to those of us lurking about that may complain once in a while - but are actually quite successful in what we do.

rbacal

8:18 pm on Nov 16, 2006 (gmt 0)



Instead of just talking down the the posters, why not offer some advice?

There's clearly little point. When people who have been successful with QS post specifics -- for example, like making sure your entire site is reputable (along with specifics for doing so), the response here is "Who the hell is google to tell me...yada yada".

The threads get hijacked, and buried in pony dung, and the people who really don't want to change just complain.

I figure at this point that if I'm going to take the time to offer specific suggestions beyond the general "here's how it works", that I'll get paid for it. (I have no interest in doing that, so...)

Keep hacking at people who are offering advice you don't want to try to understand and hear and you'll just get the blissfully ignorantly disenchanted advising the less blissfully ignorant.

Reminds me of another webmaster forum where at least 50-60% of the messages are so obviously wrong factually, that you got to wonder whether people are INTENTIONALLY giving others bad information.

exmoorbeast

9:14 pm on Nov 16, 2006 (gmt 0)

10+ Year Member



Ok kids...peace all round.

Anyone care to comment on this:

"All the latest Quality Score adjustments reward advertisers whose prospects don't click the BACK button and do another search!

Read it on some email list, thought it might ease things here in court!

mimmo

10:07 pm on Nov 16, 2006 (gmt 0)

10+ Year Member



All the latest Quality Score adjustments reward advertisers whose prospects don't click the BACK button and do another search!

I do not believe this can be true because:
- Most people check out 3/4 ads always
- The ad in first position is the one getting most 'back' clicks
- Users will always click 'back', even if they are satisfied with the AdWords land page they visited

etc.

Pengi

10:16 pm on Nov 16, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



All good points mimmo

But what if the alorithm hits sites that have a high proportion of "Back Button" hits within seconds of arrival? I know if I'm searching, there are sites I instantly reject - maybe some are quality sites, in which case they would have lots of other visitors who stay a while. But sites that have, say a 50% rate of visitors clicking the back button within 2 seconds would probably not be very good.

pdivi

10:17 pm on Nov 16, 2006 (gmt 0)

10+ Year Member



Seems the back button thing would almost be a given if G were trying to measure relevance and LP quality. Who doesn't measure their site's time/visit and pages/visit? Those are core metrics from as far back as I can remember.

venrooy

10:22 pm on Nov 16, 2006 (gmt 0)

10+ Year Member



Keep hacking at people who are offering advice you don't want to try to understand and hear and you'll just get the blissfully ignorantly disenchanted advising the less blissfully ignorant.

Are you trying to claim that you've given advice in this thread? Not only are you calling us failures - but now you think that we don't know how to read. If you're so convinced that your criticism is falling on deaf ears - then why don't you just keep your opinions to yourself? Or do you enjoy wasting your own time?

venrooy

10:27 pm on Nov 16, 2006 (gmt 0)

10+ Year Member



Back to the subject- it would be easy to set up a 2 or 3 second flash that would open up a new page at the end. Flash scripting is fairly undetectable. That would prevent someone from so easily backing out of a page. But I'm not sure google measures that.

Pengi

10:50 pm on Nov 16, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Keep hacking at people who are offering advice you don't want to try to understand and hear and you'll just get the blissfully ignorantly disenchanted advising the less blissfully ignorant.


Are you trying to claim that you've given advice in this thread? Not only are you calling us failures - but now you think that we don't know how to read. If you're so convinced that your criticism is falling on deaf ears - then why don't you just keep your opinions to yourself? Or do you enjoy wasting your own time?

Perhaps not in this thread, but in similar threads from earlier rounds of the QS I have seen R provide a gread deal of advice based on his view that rigorous compliance to the TOS and provision of a site that provides value to the surfer is effective at avoiding the QS.

Nevertheless - prehaps the frustration comes through sometimes ;)

venrooy

11:12 pm on Nov 16, 2006 (gmt 0)

10+ Year Member



Nevertheless - prehaps the frustration comes through sometimes ;)

I'm guilty of letting my frustrations show as well - So I don't blame him for that. :)

mimmo

11:49 pm on Nov 16, 2006 (gmt 0)

10+ Year Member



Seems the back button thing would almost be a given if G were trying to measure relevance and LP quality. Who doesn't measure their site's time/visit and pages/visit? Those are core metrics from as far back as I can remember.

Measuring the 'back' button is not easy... according to my programming experience when you click 'back' on the browser, the browser does not actually request the page again to the server, it just shows you what you have in the cache.

My gut feeling is that tracking the back button would be too complex and not very useful for Google.

Regarding the amount of time you spend on the landing page ... also the type of connection you have and the amount it takes to load the page would be a factor....

Anyway! The solution is to have a page that is relevant to your ads, that is in your own interest, unless you like to to throw away money :-)

Philosopher

11:57 pm on Nov 16, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Well, it wouldn't necessarily be the Back button that was tracked, but when a browser of a given IP with a given cookie clicked on another ad after the first.

That would be very easy to track.

mimmo

12:01 am on Nov 17, 2006 (gmt 0)

10+ Year Member



needs cookies activated ...
needs static IP ...
slowly loading landpages are better quality?

mimmo

12:08 am on Nov 17, 2006 (gmt 0)

10+ Year Member



just thought about something that may be more useful for Google ...
as you suggest, what if Google tracks some users with a IP/cookie combination to cross analyse AdWords and AdSense ... I mean you could identify very easily all the MFA becuase after an AdWords click always come a AdSense click ... I think analysing AdWords , AdSense click chains can be done statistically by Google and if you are in the middle of a chain you are actaully a very low quality site ... so this would have nothing to do with how a page is built, it would be all about user clicks.

Another possibility, with this type of tracking, is that if after clicking on AdWords you end up on page A and then on page B and page B is featured on Adwords for the same keyword, page A would be low quality as well... so do not have links to other pages featured on adwords on your landing pages.

xdude

2:06 am on Nov 17, 2006 (gmt 0)

10+ Year Member



Rbacal, perhaps you should just let people provide negative feedback or even rant. You've been diligently preaching that "one should either understand what Google wants and do whatever necessary to please Google, or go home and live shamefully thereafer". Not everyone silently bows to power and submits to everything crushed upon him or her. Some work hard to improve their ads/websites according to vague guidelines but are still punished by Google, hence complain. Some choose not to conform to unreasonable, changeable, contradictory rules, and complain. Some choose to simply move to other ad service providers. Yes, sooner or later other providers will also factor in ad quality so it's better to learn and cope with it now in Google's way. But I think what many of us have problem with is that this Google's-way quality has little to nothing to do with quality.

I can tell MSN has already factored in some sorts of ad quality but at least they aren't trying to bully smaller advertisers with bogus quality slogan. To me MSN still has professional respect for their customers. Yahoo is going to have the quality ingredient in their upcoming upgrade. But I bet they will most likely conduct it in similar fashion as MSN's. I believe Microsoft and Yahoo have a deeper root in the business world, hence more professionalism in treating their customers. I think Google still has the brightest brains in the technological business world. I also think they really wanted to come up with better quality in their free/paid search results. They just didn't do it right this time - besides showing a lack of professional respect to their customers, they superficially rate ad quality with biased emphasis on popularity-based ranking, but without the most important factor, sales conversion. Hmmm, please no more "ad is not (or not all) about selling".

rbacal

2:38 am on Nov 17, 2006 (gmt 0)



But I think what many of us have problem with is that this Google's-way quality has little to nothing to do with quality.

Personally, I'm in business and have no interest in complaining or "evaluating" what google does. So, I have a suggestion.

If you want to complain, start a thread labelled "google complaints" or something, and allow those of us that are more interested in learning how to "do good business" to converse without having you, or those that want to complain, hijack those conversations.

I'd be quite happy promising NOT to hijack your "I want to complain" threads if you and others that want to complain, stop hijacking threads that might help people a) understand how google works, and b) build and maintain sites that will generate revenue, not just this week, but over years.

If I say, "do x, y and z" to help yourself, EVERY single time, the thread gets hijacked by someone who wants to complain and says "But whyyyyyyyyyy do I have to do x, y and z?, I don't wannnnnna. Who is google to tell me what to dooooo?"

and on and on. What's the point?

venrooy

3:21 am on Nov 17, 2006 (gmt 0)

10+ Year Member



I don't think that we need to label the threads - anyone with any reading comprehension could tell by the first post on this thread that it was a complaint thread.

A very legitimate and not easily explainable complaint in my opinion.

2 separate people promoting the same exact webpage - with the same exact keyword. How can they have 2 separate Quality scores?

There are some explanations and they are worth exploring - And I find it unnecessary to belittle anyone that doesn't happen to agree with your explanation.

One explanation might be that Google keeps a history of performance for each keyword, and the keyword may have performed more poorly for you than for your affiliate - based on the ad verbiage you used. Do you and your affiliate use the same ads as well?

This may explain why some can move their page to a new domain, and start off with a better QS.

There are many possible explanations - I don't think that you hold the monoploly on right answers.

holyearth

3:30 am on Nov 17, 2006 (gmt 0)

10+ Year Member



I agree with Venrooy.

There is much more to this QS crap that we haven't been told.

I know for a fact that certain "Fortune 500" domains are EXEMPT from any QS. While other "Fortune 500" domains have a minimum bid even if the QS is the highest it could be.

Speaking from the advertising dollars that Google has lost from me, I have to assume that this has to hurt their bottom line.

Google has such a large piece of the search pie that they can do with us as they please. We have to sit back and watch.

rbacal

3:39 am on Nov 17, 2006 (gmt 0)



2 separate people promoting the same exact webpage - with the same exact keyword. How can they have 2 separate Quality scores?

Easy easy question. QS takes into account a number of variables associated with profiles created in part by people, then translated into an algo. The people classify a sample of sites as "quality" or not, creating two piles. THEN, through analysis, they do something similar to a discrimminant analysis, which means identifying quantifiable variables that DISCRIMINATE between the site in the two different piles.

OK so far?

That's the preamble. Now, QS takes into account a number of variables, both on the landing page, and connected with THE SITE (and other variables found to discriminate).

No two webpages will have identical values on all variables. One might be created on a site registered two years ago, another 2 months ago. One might be located in one country and the other another. One may have hidden registration and not the other. One may rank at 101 for a keyword in google and the other not rank at all. One may be seen as a duplicate of the other.

...and on and on.

So, to the algo even though the webpages are "identical", the variable used to calculate qs are not.

Think about search engine results. Two identical pages. One ranks in the top ten for a keyword set. The other, seen as a duplicate (because it is) ranks at 10,000,000. Identical pages. Same idea. Same rationale -- meeting the needs of the end user.

It's a bit more complex than that when it gets to the actual formulas that differentially weight the variables used to calculate QS, because the weights for each variable in the formula are altered depending on the values of the other variables.

It also means that two sites with identical QS's (assuming we knew the numbers) could get them in completely different ways.

Once you "get" all this, then you realize that google has provided incredibly clear guidelines to avoid QS penalties, and you also realize you can't tweak around the edges (change one thing) and hope to succeed over time. You also realize that what will work for one page on one site won't probably work on even an INDENTICAL page on another site.

If you want to know why google won't give you a "change this variable" and you'll be fine", it's because they would be misleading you by doing so. It doesn't work that way. What DOES work is to create good, non-duplicate, non-feed sites that add unique value to real people.

venrooy

4:05 am on Nov 17, 2006 (gmt 0)

10+ Year Member



That explanation is all well and good - assuming that we are talking about 2 identical websites. But we are not.

We are talking about the same exact website being promoted by 2 separate people. And therefore 90% of your explanation now goes into the southward bound hand-basket.

Again I say - You do not hold the monopoly on right answers.

rbacal

4:16 am on Nov 17, 2006 (gmt 0)



That explanation is all well and good - assuming that we are talking about 2 identical websites. But we are not.

We are talking about the same exact website being promoted by 2 separate people. And therefore 90% of your explanation now goes into the southward bound hand-basket.

This is why I get frustrated.

It doesn't change anything. READ what I wrote, and think, for George's sake.

Were the ad campaigns created on the same day? What's the ad history of each of the two people? Are the ads identical? Bids? Bid history for the two people? Do those two people have other websites? Or promote different other programs? The variables involved will NOT be the same. Look and think below the surface, ya?

Just the friggin fact that they are DUPLICATES means they will be treated differently, just as in the case of search results. DO you need clarification on that point?

ajwebmaster

4:22 am on Nov 17, 2006 (gmt 0)

10+ Year Member



My rep told me today to essentially take my money elsewhere....shocking but at least an honest answer... I would love for the stock analyst to have heard our conversation...

venrooy

4:28 am on Nov 17, 2006 (gmt 0)

10+ Year Member



This is why I get frustrated. You asked about two identical webpages, not two identical websites. Sorry, but if you want me to walk you through something, I can't read your mind.

Actually I think my words were "The same exact webpage" not "two identical webpages". That would imply same site - url - etc.

Just the friggin fact that they are DUPLICATES means they will be treated differently, just as in the case of search results.

Again - I don't think you're getting the picture - This is the Same exact web page - on the same exact website - owned and registered by the same exact person - but promoted by 2 separate people - who are receiving 2 separate quality scores.

venrooy

4:34 am on Nov 17, 2006 (gmt 0)

10+ Year Member



Were the ad campaigns created on the same day? What's the ad history of each of the two people? Are the ads identical? Bids? Bid history for the two people? Do those two people have other websites? Or promote different other programs? The variables involved will NOT be the same. Look and think below the surface, ya?

Wow that's almost exactly the explanation I gave above - right before you gave your "that question is easy" explanation.

Hmmmm...

Why don't you just admit that you are pulling these explanations out of your netheregions as you go along?

mimmo

4:35 am on Nov 17, 2006 (gmt 0)

10+ Year Member



Then this proves that the landing page itself is not the only factor in the QS equation. Google probably includes also visitor statistics, to and from user's clicks, and other things that were statistically different for the same landing page promoted by different people and visited by different people.
Google can use info from the Google toolbar, AdWords, AdSense, Google Analytics, etc. It may all be combined.

rbacal

4:39 am on Nov 17, 2006 (gmt 0)




Then this proves that the landing page itself is not the only factor in the QS equation

EXACTLY! Google's help files on QS state that landing page is not the only factor.

The information is all public, available, and clear, but you have to READ it and understand how google works to understand it.

ajwebmaster

4:41 am on Nov 17, 2006 (gmt 0)

10+ Year Member



for many the message is go away - you dont make the cut - but it is wrapped up in QS this and that and mumbo jumbo to keep you from getting your feelings hurt - no one likes to be told that their baby is ugly...
This 65 message thread spans 3 pages: 65