Adwords Quality Update II = Price Gouging Round II

29 comments

Since I was one of the "small number" of advertisers affected by the "quality update" back in July I had hoped I wouldn't be affected when they announced another "quality score" update announced earlier this week. Looking at my data (published below) I can say these updates aren't about "quality" at all. Instead they are about jacking up the prices and pricing out the riff raff.

adwords-export2

The data published in the screen shot above is yeasterdays data, the keywords have been sanitized, but other than that the data is unaltered.

Looking at keyword #1, this keyword is now disabled for low quality and I need to raise my bid to $0.40 to activate it. Looking a little deeper we see it's running in position 4.1 with an 11% CTR and a 41% conversion ratio. So my ad is compelling enough to get clicks, and delivers what it promises since 41% convert, wouldn't fit my definition of "low quality" but don't forget to look at the average CPC price of $0.33.

Keyword # 2 running in position 4.5 with a 7% CTR and a 71% conversion. That keyword seems ok but take note of the higher CPC of $0.52.

Keyword #3 running in position 8.1 witha 2% CTR and a 0% conversion. Now if any keyword was to b flagged for low quality this should be it, however look at that CPC at $0.47.

Now the your next question might be well what are the keywords and are the related? They related in the way "soda", "cola", "pop", and "soft drink" are. The next question you may ask is what about the landing page, well I can tell you they all go to the exact same landing page.

Looking at all of the keywords for this site I was only able to find one below $0.40 per CPC running at $0.38. Is Google rigging the game? Are they so dominant that they can raise prices arbitrarily under the guise "quality improvements" with impunity and without fear of repercussions. Should they be allowed to? With the ability to "give themselves a raise" whenever they want is it any surprise they have profits every quarter while others do poorly.

Comments

Great Stuff!

This is great stuff... AT&T did crap like that in 1984 ... so did Standard Oil in 1910..

The point is.. when you reach near monoploy status.. this happens... and no one is there to stop it.

MSN, Yahoo, Ask, Clutsy are lightyears behind Google in marketshare... hence it gives them a free hand to dispence any type of 'justice' they see fit.

...or...

Could it be that, as solid as you numbers seem, maybe your competitors do a better job? I mean you have your numbers to compare against, but G has everyone's numbers.

Maybe a 41% conversion rate is low, relatively speaking? (I personally can't see how, but...)

My point is simply that seeing your numbers, while handy, doesn't give a true picture of the actual market. Sadly, only G sees that - and they aren't sharing.

I'm not trying to say G is good, blah, blah, blah - just that we really can't draw any conclusions on this from your data.

Still, even though I'm trying for balance here, I'm with you on not seeing how those sorts of numbers would be classed as "poor performing"...

There's a serious lack of transparency

and it is worrying. Google can pretty much do as they please without any kind of reprecussion all in the name of landing page quality algo. How are we ever really going to see the full picture with everyones stats? We're not.

Don't forget - the likes of Yahoo are moving towards a similar technique aswell for early next year.

Andrew Goodman wrote a piece on it a while back http://www.traffick.com/2006/11/lack-of-editorial-transparency-nagging.asp

Bot Masking

OH and another thing they are checking the pages with a new bot that doesn't use the Googlebot or Adwords Bot User agent

72.14.194.32 - - [09/Nov/2006:10:00:58 -0700] "GET /XXXXXXX HTTP/1.1" 200 5033 "-" "cks"

One thing I noticed about

One thing I noticed about this is that the increase in minimum bids ( for me ) was much less marked than it was back in July. Also the number of phrases affected ( as a percentage of all my phrases ) seems to be less than it was in July.

makes no sense

>>maybe your competitors do a better job?

mayby they do although 10% CTR and 40% conversion is pretty good, but if that's the case then the ad should simply be dropped down the list. If the ads positioned 5, 6 and 7 perform better then that then I can understand the quality score indicating that the position should drop, but I've never understood how paying more can override a quality score if this is 'for the user' - as far as the users concerned the page is good or bad, its not better if you pay twice as much.

Disabling something with a high conversion rate or attempting to price it out of the adspace is surely detrimental to everyone?

seems like they are going

seems like they are going after money phrases... ie: phrases that have a decent amount of clicks that they think you will still run and bump the bids a bit.

$ GRAB.

mayby they do although 10%

mayby they do although 10% CTR and 40% conversion is pretty good, but if that's the case then the ad should simply be dropped down the list. If the ads positioned 5, 6 and 7 perform better then that then I can understand the quality score indicating that the position should drop, but I've never understood how paying more can override a quality score if this is 'for the user' - as far as the users concerned the page is good or bad, its not better if you pay twice as much.

The landing page quality doesn't effect your ad ranking, its only directly correlated with your minimum cpc. The rest of the factors that make up quality score effect your ad rank. Perhaps this is where they are going wrong to exclude it as something seperate?.

I can't see how a keyword with a 11% CTR and 41% conversion rate can be hit so badly when its obviously performing. If your competitors are in fact performing better then why not just let QS/bid determine your rank rather than throw in a min cpc.

"I can't see how a keyword

"I can't see how a keyword with a 11% CTR and 41% conversion rate can be hit so badly when its obviously performing."

A less generous interpretation would be that the keyword is obviously performing, and Google knows it. An interesting experiment would be to run equivalent pages and bids on two different sites, one with Google conversion tracking and one without and see if there are any differences in minimum bid amounts.

You mean

you told them your conversion rate?!!!

Plenty of people have been paranoid that if Google say conversion rates, Google would push up the prices on things that work. It's not hard to think that if they see a term getting lots of clicks (number 1 pulled twice number 2) and a high conversion and a low CPC, up the CPC.

Not saying this is happening, but with even less transparency, I think it gives people even less reason to want Google to see what their conversion rates off. Might be interesting to track your conversion outside their system and see if that stops those type of changes.

We got banged too

Funny with us it is not conversion but actual quality ... we have been sweating down the bids for very specific keywords that land on a definition page for the term - now where is the downside?
Well we have AdSense on there so they must think we can afford to pay more.
Like you we are getting 5-10+% CTR and the ads are sitting at positions 4 and 5 - though even ones where there are only 2 advertisers are getting pushed.
We had 600 plus words banged on Tuesday.

Absolute power corrupts absolutely

Funny how the QS changes seem to sync with quarterly P&L announcments. Since both sets of QS changes impact only pricing (not ad rank), they seem a clear attempt to squeeze advertisers' margins.

Thanks to the conversion data that Google Analytics (which includes sale value) and AdSense users provide, Google even know exactly how much of a squeeze they can put on a user.

Every AdWords account specifies which industry the user operates in (eg, automotive, affiliate marketing etc). Google could use this ROI data to calculate industry averages, allowing them to decide how much of a squeeze to put on a user by the sector they operate in.

Some industries (eg, affiliate marketing) will have far greater price elasticity for AdWords than, say, mom 'n' pop e-commerce sites.

Since such a high proportion of Google's revenue comes from AdWords/Adsense, even small tweaks to minimum bids etc could scale to substantial profit increases.

Conversion Testing

Actually the campaigns with conversions did better on this quality update than those without (oops was I not supposed to test that!)

I mean it's not like they might be using user generated data ...

I'm really glad...

that we all used the "free" conversion tracking. *cough* I'm sure the CPA models G will inevitably roll out will LOWER our cost per acquisition prices too:)

quality schmality

Quote:
The landing page quality doesn't effect your ad ranking, its only directly correlated with your minimum cpc. The rest of the factors that make up quality score effect your ad rank. Perhaps this is where they are going wrong to exclude it as something seperate?.

totally - and its so fundamental to the idea of 'showing quality ads' that I can only assume that they worked out what to do to increase profits and then tried to think of an reason for doing it. Quality didn't enter into it until meeting 4 when someone suggested it might be a good excuse.

If you hsve an ad which is misleading/leads to a horrible landing page/doesn't convert then you can improve the user experience by either stopping the ad running, let it run but with a penalty so its less visible, or help/get the webmaster to improve the page.

So they don't drop it down the page, they don't tell you whats wrong so you can improve it, and they stop it but restart it with no comment if you pay more. There's no star next to it to help the user by saying "*This ad is a bit shit but they're paying more so we're letting the run it" and no feedback to the website owner, so how is anything, except Googles bank balance, being improved for anyone?

I guess the argument would be that crappy ads stop being run because they cost too much to run - but why go through that pantomime when you can just stop the ad and refuse to restart it?

They need a few people with some common sense and business acumen as well as ones with IT PhD.s imho

Why shouldn't they be

Why shouldn't they be allowed to raise their price whenever they want? Surely you're not suggesting that the government become involved and start setting prices for PPC?

tell me you're trolling

before I have to replaster this wall I'm banging my head against.....

Don't be evil. Oh, if you *insist*.

Peter, price fixing made possible because there is no viable competition in a market is an abuse of monopoly power.

Not to mention a shabby way to treat the people who pay to keep the Googleplex stocked with fussball tables (ie, Google's advertisers).

Graywolf, interesting to hear that your campaigns with conversion tracking were hit less hard. Any other notable differences between what did and didn't get hit you'd care to share?

flip side

and lets look at the flip side, with publisher hat on.
Across my sites, I've not seen any changes in advertisers, nor an increase in CPC/ECPM.
Either my advertisers are all Really High Quality already! (*cough*) or the cost increases aren't being passed on to the publisher.
Anyone else seeing their ECPM or CPC increase?

I'm with Danny

...in that I've often thought that giving Google a peek at my conversion rates might not be a good idea. A keyword with above average/phenomenal conversion rates would be ripe for a forced CPC increase, because Google can determine that the advertiser is making money, and thus will likely comply with Google's demand to raise their ad spend.

Just because I'm paranoid doesn't mean they're not out to get me (well, my money).

What's the score?

How can you tell if your landing page is crap or you're just paying the going rate?

I mean, If I'm bidding on a new term or keyword and I've got no history, i.e. I didn't know the price of a keyword before these changes, how do I know I could get clicks cheaper if I improved my landing / page site?

eCPM

Anyone else seeing their ECPM or CPC increase?

Yes. Up about 40% on two sites.

I'm curious to know if this

I'm curious to know if this is for everything or just competitive key word sets.

I think they might be doing this to certain categorys and not others.

Something to Consider...

So, here's a tidbit you might like to hear. I was expanding a certain account the other day and pulling adgroups out of a large campaign and setting them up as their own campaigns. In one of the new campaigns/adgroups all of the keywords were flagged and went inactive. I thought that was pretty odd since I had copied them directly from an adgroup which was still live and had even raised the bids a little for the new set. Perhaps it was this new update. Whatever.

Anyway, I figured that I would see if I could get around the quality score pay-to-play scheme by doing something other than upping the bids (they wanted me to go from $0.30 to $5.00 so I had some good incentive to try). Here's what I did:

First off, I gave it some time. The new campaign was setup in the afternoon and I waited until the next morning to work on it (not sure if this mattered at all, but it's what I did). Next, I edited the ad copy. There were 2 ads running, 1 of which was not using keyword insertion in the title. I added keyword insertion to the title. The second ad did not mention the brand name of the product I was advertising in the description (but it had keyword insertion) so I added that. Then I waited some more.

It worked. By the next morning my ads were running just fine with a $0.30 bid. So, before you pay, I would suggest that you make some changes and give the system a little time to digest them. You never know, you might get lucky.

you told them your

Quote:
you told them your conversion rate?!!! Plenty of people have been paranoid that if Google say conversion rates, Google would push up the prices on things that work. It's not hard to think that if they see a term getting lots of clicks (number 1 pulled twice number 2) and a high conversion and a low CPC, up the CPC.

Hey I'm liking this Danny guy ;-)

The occasional bug

I had one keyword, in a campaign in Europe, where there were no other bidders for the keyword.
I thought, cool, i can get this one cheap. Other keywords in that language were such that i thought i would get it for about 10 cents american.
Entered the keyword and it said, minimum bid is 5 bucks american.
So I go to google and type in that keyword, and ...nothing, no ads.
And then, i looked later, and i had one impression, where yep, my one impression came in in average position 1.0.
Thats gotta be a bug, I guess? Why would the minimum bid price out all bidders?
Ron Drabkin
www.adisem.com

Cringely

Hmmm... Cringely covered Adwords late last year, and I am sure there was mention of someone with an obscure low-market mathematical product who was expected to pay over $1 per click despite there being no competition. IIRC this was down to low ctr.
Of course, despite searching, I can't find the article now :(

But, basically - its Google's way of making sure their near-monopoly makes lots of money. What, you are surprised? :(

date range and bot

what are the stats for a longer date range for your selected keywords. Using "Yesterday" doesn't always tell the whole story.

Has anyone tried blocking the bot with a robots.txt file? if so what happens?

bugs and greed

There have been bugs that just don't show you - we have had that for quite some time before Google admitted it was a bug... but a lot of what is happening is inactivating and upping bid requirements.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.