Welcome to WebmasterWorld Guest from

Forum Moderators: buckworks & eWhisper & skibum

Message Too Old, No Replies

Ad Quality Change

Fewer ads on some queries and more on others


running scared

9:55 am on Apr 27, 2006 (gmt 0)

10+ Year Member

From Inside AdWords - [adwords.blogspot.com...]

Starting today, and over the coming weeks, we'll be implementing an ads quality change designed to show fewer ads on queries for which our users might prefer not to see them and more ads on queries for which ads are useful.

Does this mean more blue real estate for commercial queries?


1:32 pm on Apr 27, 2006 (gmt 0)

10+ Year Member

"With this in mind, on queries such as 'car insurance' or 'flowers,' we try to show a wide selection of relevant ads to our users. And, for queries where users are most likely not interested in a product or service, such as 'dog friendly parks in Mountain View,' we may aim to show fewer ads."

It looks like the ads are going to be more targetted, but I bet more expensive as we fight to be show

p.s I wanted to show this quote in a grey box - how would I have done that? - thanks

running scared

6:29 pm on Apr 27, 2006 (gmt 0)

10+ Year Member

Does nobody else think this is interesting?

Issues like broad and phrase match becoming even more "broken"? The apparent contradiction of encouraging people to add important terms, whilst encouraging us to reduce the number of phrases? Any hope of this de-broadening of broad match being applied to the content network (where surely it is needed)?

If ever we wanted evidence of G doing it's best to make AdWords the users' results of choice for commercial queries, then surely this must add to the jigsaw? (Not that I have a problem with that)

(Web publisher - have stickied you the grey quote thing)


7:27 pm on Apr 27, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

Yes, I think this is *very* interesting, perhaps the most interesting AdWords-related announcement from Google since, oh, mid-August.

With regards to your statement about G wanting users to increasingly think of AdWords as the results-of-choice for commercial queries, think about the phone book. Do the yellow pages firms want you looking businesses up in the white, small text business listings, or the yellow pages? And as a user, which do you end up using, and for what % of your searches?

I'd say AdWords could be getting 60-70% of commercial clicks at some point in the not-too-distant future.



11:36 am on Apr 28, 2006 (gmt 0)

10+ Year Member

It will be interesting to see if they mention this major change anywhere else besides their blog...


12:48 pm on Apr 28, 2006 (gmt 0)

10+ Year Member

When Adwords was first launched relevency requirements were simple and well articulated to the advertisers.

An ad had to achieve a 1/2 % CTR minimum on the Google network, and the ad had to accurately describe the website it was taking the searcher too. If the ad/url was inappropriate or the CTR was too low, the ad was disabled. The searcher was spared having to see that ad. And the advertiser understood why his ad didn't fit in.

Many of the individual steps to improve ad quality since then seem logical when taken separately. But somehow the system overall has evolved to a point where publishers complain about irrelevent ads, Google is planning yet another change based on the fact that they believe searchers are being inundated with ads that are degrading the searchers experience, for non-commercial terms. And advertisers (at least many who post on this board) are constantly baffled by what is happening.

It seems to me that going back to the simple rules of CTR and ad/url combination relevence would result in better searcher and a better advertiser experience. I really believe that if Google were to make searchers and advertisers happier, somehow they would maker more money.

Lets take the blogspot example non-commercial search "dog friendly parks in Mountain View"

Hypothetically lets say the worlds best ice cream cart roams various dog friendly parks in mountain view (vendor likes to bring his dog). The ice cream vendor offers online coupons for his ice cream cones. So he advertises on the search phrase "dog friendly parks in Mountain View" with and ad that says

"print out a coupon for 30% off abc ice cream cones anywhere in mountain view"

Under the old system as long as 1/2% of Google searchers decided that was relevent the ad would would run. If the ad did not achieve the required ctr, it would become disabled searchers would be spared the ad, and the advertiser would know exactly why.

If the new system is another twist on quality score, the robot that reads things wouldn't see any mention of dogs or parks in the ad or the landing page and set a high minimum cpc to appear even if the ad is getting a 10% CTR. The vendor won't be able to afford the new higher minimum, and the ad won't run.

After contacting Google customer support the advertiser will recieve several vague boiler plate answers to his question "why do i have to pay $5.00 per click to have a coupon offer for a $2.00 ice cream cone?" If he continues to be persistant in seeking an answer he will eventually be told that his question will not be answered for risk of comprising the integrity of the "algorithm".

As for the 10% of the searchers who though that ad was relevent enough to click, and who also got to save $.60 on their delicious ice cream cone, they will remain the lucky ones. Future Google searchers for "dog friendly parks in Mountain View" will sadly never have the chance to buy a delicious discount ice cream cone.

This example is a long winded way of saying that the marketplace over the long run will determine better the ad relevency than any quality score algorithm can hope to be. If 1/2% on the Google network was too low for relency, then raise the percentage. If 1/2% is appropriate for some keywords, but not others than assign a different ctr for those keywords and let the advertiser know what they bogey is.

The current lack of transparecy to the advertiser combined with the ability for a well funded advertiser interested only in branding to force irrelevent ads into the system by paying high CPCs seems to me to be far less optimal than the original transparent system that disabled irrelevent ads for being, well, irrelevent.


5:18 pm on Apr 28, 2006 (gmt 0)

10+ Year Member

Nicely summarised Jim2003. So reading between the lines basically what google is saying is "We're going to show more and more ads with high CPC that will make us good money on some searches, and to justify it as an enhancement for the user we'll remove one or two ads that won't make us much money anyway on other searches"



5:21 pm on Apr 28, 2006 (gmt 0)

10+ Year Member

fewer ads on queries for which our users might prefer not to see them and more ads on queries for which ads are useful.

Seems a bit contradictory to me, whether users would prefer to see ads or not is not the same thing as whether those ads are useful.



7:34 pm on Apr 28, 2006 (gmt 0)

10+ Year Member

This change wouldn't necessarily make it harder to display the ice cream cone ad. That advertiser will be trying to get their ad displayed in a smaller universe -- say no more than 5 ads showing, instead of 11. But, the change will also have a systemic impact on the overall level of competition for these types of obscure searches.

Some of the big players may drop out of the market, or change strategies, because of this change.

On balance, the chances of displaying the ice cream cone ad might actually improve, if the new system weeds out high volume advertisers that run auto-generated ads on tens of thousands of low cost "longtail" search terms.

Regardless of whether it helps or hurts particular types of advertisers, I suspect this will help google's bottom line. Reducing the number of ads on relatively non-competitive searches may intensify bidding for the small number of remaining slots.

In effect, google is creating artificial scarcity in markets where every player can currently run their ad without getting into a bidding war.

Google will simultaneously be expanding its inventory in other markets, where there is no lack of participants -- just a lack of sufficient ad space to sell. By creating more ad space for certain "commercial" searches, google can increase ther atio of revenue-generating clicks relative to "free" clicks. If this is done selectively, it may increase total paid clicks without reducing the amounts paid per click for the most desirable ad positions.


9:23 pm on Apr 28, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

When Adwords was first launched relevency requirements were simple and well articulated to the advertisers.

Jim - I hear your point. But, I recall little that was simple or all that well articulated about the previous system of disabled keywords. It, in my memory, was never as black and white with regards to CTR as the "rules" stated... many words were disabled with CTRs that far surpassed oft-quoted floor requirements or seemingly without enough traffic to have made a statistically valid determination.

Not sure the impact of these latest changes, but I do feel - despite gripes about the increasing CPCs required to enable keywords, IMHO, the system has become simpler and more transparent over time.


7:21 pm on Apr 29, 2006 (gmt 0)

10+ Year Member

i think econman hit the nail on the head with "artificial scarcity"

for a commercial term there are lots of advertisers, who are bidding high dollar amounts. google knows that most visitors probably click on the top 5 or 10 ads, which are all relatively high paying, so google is happy.

for non-commercial terms, however, there might be 20 advertisers, but they are all bidding low amounts, say between 5 and 10 cents. so, with this change, google limits the playing field for this search to say, 5 advertisers. the other 15 are now out of the running, unless they bid more. if they do bid more, then google is happy because they have just raised the value of that search term, if not, then google only loses some low-bidding advertisements, and the user has fewer ads to wade through.


9:33 pm on Apr 29, 2006 (gmt 0)

10+ Year Member


All good points. If the system works as you are imagining. But the quality score system for landing pages seems to have serious deficiencies. Landing pages that are keyword stuffed with "dog friendly parks in moutain view" and using dynaminc keywords in the ad will be rewarded. And the ice cream page may never get started because there is no mention of dogs, friendly, mountain view or parks on the web page.


9:30 pm on Apr 30, 2006 (gmt 0)

10+ Year Member

Let's talk bottom line.

Google is making a big mistake trying to assume they know what people want to see. Their arrogance is starting to show.

My Adwords expenditures over 1 week:

4/22/2006 - 264,629 impressions, 5,203 clicks, $703 spent

4/29/2006 - 163,578 impressions, 2,864 clicks, $378 spent

I didn't change anything - Google did.

If they only want 1/2 of my money, so be it. I'm not going to spend the time trying to figure out their riduclous system that changes every 3 months.

People (and advertisers) like consistency, not inconsistency.

Google's stock will plumment, as they rely on Adwords for 95% of their revenue.


10:20 pm on Apr 30, 2006 (gmt 0)

10+ Year Member

I agree with most of the posts here...

It seems that the last few changes that G has taken, and those in planning benefit adwords greatly in terms of earnings.

As advertisers, we do love consistency. It seems that Adwords has too much change associated with it. And, this change is mostly negative.

What did we expect? Their stock's gotta keep it's energy somehow. Adwords accounts for 90%+ of G's moolah. Can't say you didn't see this coming. To add insult to injury, I see this getting worse (like gas prices)...

When will it stop is the ultimate question.


4:35 pm on May 1, 2006 (gmt 0)

WebmasterWorld Senior Member jtara is a WebmasterWorld Top Contributor of All Time 5+ Year Member

Some people at Google need to take off their tin foil hats for long enough to enjoy a tasty ice cream cone...

Which is to say, that jim2003 has perfectly articulated the "disconnect with reality" at Google that I've harped on here.

There's the real world. And there's the Google world. And precious little intersection.


5:43 am on May 5, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

There's the real world. And there's the Google world. And precious little intersection.

Makes me wonder what world some of these posters are living in. Last time I checked googles market share has been increasing. This sounds more like a warning shot at ebay.


1:15 pm on May 5, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

I personally cannot tell if the changes at Google were good or bad for the searcher. But Google probably can. They know over time if their changes are successful.

My biggest complaint is the constant change. It never stops. We frequently end our meetings with, "We hope we make it through the year." One day they will make a change that puts me on here after having lost 80 of my impressions, and I am trying to figure out why. Or my cost is sky high. Then I will have to do something else for a living.

Quantam Goose

1:16 pm on May 5, 2006 (gmt 0)

5+ Year Member

Googles Market Share? Ever increasing? OK, lets go down memory lane.

1980 IBM massively controls all corporate computing seemingly unchallenged with their "big Box" mainframe systems.

Lotus - the 90% dominant spreadhseet software

Wordperfect - the standard for all corporate documents.

Wang - the undisputed leader in corporate word processing systems

Netscape - the internet standard for Browsers.

On and on. Large public companies (Google being large) when tied to quarterly earnings reports BY DEFINITION lose their flexibility. They can only deal in the short term advantage to satisfy the street. Regardles of the specifics of complaints seen on this forum, they all have one common denominator. They are outcries over Google tweaking algorithms for short term gain.


4:20 pm on May 16, 2006 (gmt 0)

10+ Year Member

Could there be a maturation of growth in number of searches. For example, we advertise a general type of merchandise ( a one word description). We use exact,phrase and broad match.

Year to Date (Thru April) impressions
EXACT MATCH 345,000 last year/ 266000 This year
avg position 2.1 last year/1.3 this year

PHRASE MATCH 1040000/last year/ 876000 this year
avg position 4.7 last year/5.0 this year

BROAD MATCH 691000/last year/536000 this year
avg position 7.4/last year/7.5 this year


4:39 pm on May 16, 2006 (gmt 0)

5+ Year Member


Just curious. That breakdown you listed, I suspect you imported Adwords data into a spreadsheet or program to get that. It's not a feature built into Adwords reporting, is it?


Just like to know all my options.



Featured Threads

Hot Threads This Week

Hot Threads This Month