|Google Explains Quality Scores In AdWords|
Google Explains Quality Scores In [googleblog.blogspot.com]AdWords
|In the auctions, advertisers enter bids that reflect how much they are willing to pay for a click on their ad -- this is called their maximum cost per click (CPC). Ads are then ordered by the product of the bid that is entered and the estimated ad quality score. People often ask why ad quality enters the formula -- isn't the bid per click enough? Why can't advertisers just buy their way to to the top ad position? To see why both components are important, let us look at a simple example. |
Suppose that two advertisers are bidding on the keyword "jet airplane." Joe's Jets is selling actual jet airplanes, while Moe's Models is selling models of jet airplanes. Since jets are expensive, Joe is willing to pay a lot per click. But not many people can afford to buy jets, so Joe won't get many clicks. Moe, by contrast, is willing to pay a lot less per click, but he will also get many more clicks.
Which ad should be listed higher in the "sponsored links" section of the search results page?
What matters in this decision is not simply an advertiser's value for a single click -– the maximum CPC that the advertiser is willing to pay -- but rather the total estimated value of showing that ad: the value per click times the number of clicks that the ad is likely to receive.
So why are quality scores important? Answer: they lead to a better auction by allowing advertisers to buy clicks, publishers to sell impressions, and users to see relevant ads.
That must have been written by a committee. 5 paragraphs, doesn't say anything, and I don't really understand what it is they're saying. Certainly they didn't prove their summary in the last paragraph. They just flew that one in out of nowhere.
|So why are quality scores important? Answer: they lead to a better auction by allowing advertisers to buy clicks, publishers to sell impressions, and users to see relevant ads. |
Oops they forgot '... and Google to earn more money' - they must have studied the ratio high CPC/low click volume vs low CPC/ high click volume and found that the latter brought more in while keeping 'all users' happy.
|What matters in this decision is not simply an advertiser's value for a single click -ï¿½ the maximum CPC that the advertiser is willing to pay -- but rather the total estimated value of showing that ad: the value per click times the number of clicks that the ad is likely to receive. |
Ok, so they are saying it is not CPC, but the product of CPC and CTR that matters. This is obvious. But what does it have to do with the quality score? I don't see an explanation of the quality score here, the thread's title is misleading ...
Well I think that it is still all about Google's power to determine how the game is played according to their rules.
In their example they don't even come out to say whose ad they show, bet it is the high priced guy until his CTR is below .8%.
Google did not make tons of cash last quarter showing the little guy paying $.05 per click on that term follow my drift?
Obviously Google is going to show the ad that makes them the most money. Quality Score = Cash Flow, whoever provides the most cash has the most quality, they can gold sprinkle that all they want, it all comes down to cash.
Was that written by a drunk person? Did I miss the explanation?
No not drunk, but a lawyer. Blows a lot of smoke but does not give anything concrete. Love for Google to clearly state in the example who's ad THEY would have shown.
|What matters in this decision is not simply an advertiser's value for a single click -– the maximum CPC that the advertiser is willing to pay -- but rather the total estimated value of showing that ad: the value per click times the number of clicks that the ad is likely to receive. |
If you read “value” to mean value to Google, not to the advertiser, it’s all pretty clear (and grammatically correct). They have figured out a way to order the paid results so they make the absolute maximum amount of money possible, and the above quote from their blog says exactly that.
I thought they did a rather good job of explaining why they implement the quality score; they make more money this way. Of course its all wrapped in nauseating Google speak suggesting this is all done for the good of everyone but Google. It’s just an algorithm for ordering paid placements, just like an algorithm for organic.
I for one am looking forward to their next post where they explain how this works. Because as a person who spends a great deal of money on Adwords, I don’t think this thing works well at all.
|In this post, we will try to describe "why" we use quality scores; a later post will go into "how," including more information about bids. |
The story unfolds.
perhaps some drunk lawyers?
Crikey, that adbot is very clever knowing about the prices of private jets vs the prices of model jets!
Wow, anyone finding themselves having with high minimum bids should now be able to rest easy knowing that Google knows more about your business than you do - who needs artificial intelligence!
But seriously that example is offensive and massively arrogant. I doubt that any of the people working there actually had a job in the real world.
We already know why quality score is used... to make more money for Google.
The problem is how it is used as it doesn't use "quality" in the traditional sense of quality being relevant.
According to Google:
|Quality Score helps ensure that only the most relevant ads appear to users on Google and the Google Network. |
This isn't true if they are applying quality score to all ads - otherwise how do ebay ads show up for so many irrelevant queries?
On Google's earnings call this afternoon a Wall St analyst asked what effects the company saw from the AdWords changes implemented early September (1st Page Bid, no more Inactive KWs, real-time Q.S. calculation), and Jonathan Rosenberg replied that Google makes 10-50 (yes, *50*) 'quality improvements' *each quarter*. Two things worth noting:
1) If they have been and continue to make 10-50 changes to AdWords every quarter - and that *was* the clear implication from Rosenberg's statement - then no wonder Google never fully explains Quality Score. They *can't*, because it's a yield optimization set of calculations that change frequently, do not necessarily hold true to what Google tells the advertiser community, and aren't, in fact, all about ad quality;
2) I've listened to virtually every Google earnings call, and each time an analyst has asked a question about monetization improvements, Google execs have responded by talking about Quality Score and 'quality improvements'. That QS and monetization are synonymous inside Google, coupled with 10-50 'quality improvements' per quarter, together means that Quality Score is not the Newtonian, relatively stable system Google tells advertisers and agencies it is, but much more like a quantum particle whose traits can only be understood in terms of probability.
Probability --> Otherwise said, Google's system is a real-time yield management system working on Google's behalf. While publisher & advertiser interests being met feeds into the system, ultimately the system's goal is yield maximization, and to that goal static explanations (like the ones we in the SEM community are always trying to get out of Google) are like straight answers on quantum entanglement = either you won't get a straight answer, or the truth will blow you away. . .
So What? --> This means that advertisers and agencies need to 'instrument' their SEM efforts with a yield optimization system - working to *their* ROI goals and optimizing to better [=ROI] data than even Google has - in order to continue to thrive. Thrive, first in the yield-drive, fuzzy, world of Google, and thrive second in the wider advertising world - radio, TV, print, banner, etc - that Google's more efficient system is bound to grow into.
IMO, we in the SEM community need to stop trying to get a static definition of Quality Score, as it will never come for the reasons stated above. Instead, we need to build our own yield maximization systems so that we know at all points in time the nature of the buying intent in the traffic we buy, and optimize our traffic buy and our sites (our businesses, ultimately) to our own goals. Do that and you can give as well as you get from Google.
this is quite misleading. all they explain is once again the performance criterion:
cpc * ctr = performance
best performing ads get to show on top, because maximal performance = maximal cash for google. everyone knows that by now.
quality score has been introduced just because performance isn't enough, that's the point they are missing here. it's all about user experience. in the example, provided the ad text isn't clear cut, most users if they click on the costly jet airplane ad, don't find what they are looking for. they don't want to buy a jet airplane but rather a model of a jet airplane. they are less likely to convert for the advertiser (which is set off for him by the much higher cost of the product).
BUT as many people are unhappy with what they find, there's a higher overall frustration level. you must factor in the after-click-experience, because people who are deterred by the percieved bad or wrong offer are a) less likely to click again b) ready to complain and tell their friends c) pulling down future clickthrough rates d) a long-term threat to googles money maker.
this is where quality score kicks in. advertisers who frustrate people with misleading ad text or a crappy landing page, in other words provide a bad browsing experience get degraded with a low quality score. they get penalized, because they draw down the network credibility and future income expectations for google.
i'm really not sure if that's what they wanted to explain us here ;)
I like shorebreak's analysis. I also get the feeling that Google is lucky they don't have an "s" in their name like Micro$oft. They are heading in that direction very quickly...
|we in the SEM community need to stop trying to get a static definition of Quality Score |
Absolute true that Google AdWords is a “living thing” today.
Lives longer than a turtle from Galapagos, cannot have a cancer like a shark, changes its shape as an ameba, and cannot be looked into because it’s a one cell organism.
And nothing bad about this above. The core of this mon(k)ey making mechanism is the fact that we all make a margin which many gratefully report in Google Analytics (why would otherwise this thing be free – there is no real free stuff except what moms and grannies make for us, from their hearts), so Google AdWords can implement their 10-50 changes correctly.
So, as long as we make money there, we’ll be running it and that will be an ongoing proof to Google that they’re doing it right, regardless of what the rest of the world thinks or says.
So much about monkeys and moneys. ;)
This QS thing has really become too tiring. Some people seem to be spending way too much time onto it.
So many times it turned that while it looked like the whole world was going to burn down if we don’t do a certain thing that day, the very next morning everything looked just fine.
|But seriously that example is offensive and massively arrogant. I doubt that any of the people working there actually had a job in the real world. |
Indeed Swanson. That explanation provided nothing to me and as far as the real world.....
Moe would be foolish to simply bid on jet airplanes if he is selling model jet airplanes. What happened to the all important concept of targeted keywords?
I'd get killed advertising on such a vague term. Especially with expanded, extended, extraordinary or whatever the latest broad match concept is. I fear broad match!
I firmly believe Google has a set of words that they consider high value regardless of context. It's indeed disingenuous to believe that the Adwords algo has the "artificial intelligence" to make an evaluation what I am selling -- corporate jets or model planes.
One of my campaigns has a massive variety of "styles" which require the use of common words combined with my product.
gold widgets - I can't advertise for less than a $10.00USD bid. Even though they differ only in color from red widgets where Google will show me fine for 20 cents. 20 cents is a reasonable, fair bid for a $20.00 product.
Same with a term like travel widgets. Can't do it even though my ads make it clear these are cheap items. miniature widgets will do fine.
Then you have the "size" issue - ad says we have "size 1 through size 5 widgets". size 1 widgets and size 5 widgets will get low bids, sizes 2, 3 and 4, I can't afford to use as keywords. The ad would look silly if I wasted precious space noting we have "size 1, size 2, size 3, size 4 and size 5 widgets". This despite the fact that the landing page makes it amply clear there is a button, description, ALT tags and appropriate META keywords and a META description indicating the sizes available.
As a quick test, I set up a dummy page that Google couldn't spider. jet widgets bids were out of the question Same page, same ad with jet widgets - same ad test, now they are green widgets - 20 cents.
So either the QS is so simple, but irrelevant that Google won't admit it or so complex that no one understands it. Either way, I suspect a lot of bugs are in that QS too.
Artificial intelligence is just not there yet.
i understand what they want to accomplish, but i still see times when their broader assumptions incorporated into judging quality mean that things that consumers would definitely want to see get blocked from the auction. i can imagine if i were in G's, i'd see this too... and in their shoes i'd probably conclude that blocking a limited number of good things (innocent bystanders getting whacked) is still better than what would happen without the quality scoring in place (too much crap).
like ad text has an appeals process (request an exception), i wish certain keywords would have an appeal's process and that exceptions could be granted. and G could keep a close eye on how things behave to make grade their ability to hand out exceptions, and furthermore, use that same data as a feedback loop to improve the algorithm that scores quality.
Moe is an idiot who won't be in business very long by spending on a term that is too generic (jet airplane) for HIS industry and needs.
I would think the end user looking for a 'model airplane' would have the word 'model' somewhere in their search term.
Joe, however, will get to spend more per click because of idiots like Moe (until Moe goes broke, and is replaced by the next Moe). In fact, he's probably also paying more per click because of other bidders like Sears, eBay, Pricegrabber, and Target - which don't generally carry, um.....jet airplanes.
I agree with everyone else that says this is just a crash grab.
That summary is not an explanation, just typical Google arrogance and shaking people for money.
I know of examples of sites that have inexplicably low quality for no apparent reason, even sites with high page rank and rankings for the keywords with low quality scores.
There should be an anti-trust lawsuit against Google. They're becoming more dangerous than Microsoft.
And has been said many times before, the ebay and Ask.com ads just make a fool out of the whole qs system logic, you can't take Google's reasoning seriously with such an unbalanced playing field, you either want to get rid of junk ads or you don't, which is it Google?
Chrisuk: Here, here
I think it's about time to put this discussion to rest. Quality Score exists to optimize G's revenues, and any positive effects Q.S. has on user experience are secondary effects and not primary goals.
I am Shorebreak, and I approve of this message.