| 6:03 pm on Sep 15, 2008 (gmt 0)|
Just got email from Adwords Blog. New QS Coming down over next few days. I think they are over playing the word "dynamic". What does that mean? The query will "dynamically" look at your account history and landing page quality and "dynamically" not display your ad?!?
| 6:36 pm on Sep 15, 2008 (gmt 0)|
I think it means it will look at your query now and then and your ad will be displayed variantly, if the QS isn't high enough.
| 8:55 pm on Sep 15, 2008 (gmt 0)|
So under this new system if a keyword does not qualify for display on the first page does that mean it will show on 2nd or subsequent pages or that it will not show anywhere?
| 9:39 pm on Sep 15, 2008 (gmt 0)|
This is my opinion so take it as that - however it is based on some actual data I have access to.
The changes we are seeing here are intended to make Google more money by showing more ads and getting higher bids.
Over the past 2 years the "quality score" system has seen an incremental development, with the concept of the "adbot" and raising minimum bids in tandem with creating a bid landscape that incorporates this "quality" into the auction.
What this really means is that Google has developed extensions to the algorithm that determines the minimum bid anyone has to pay on a particular keyword to send that traffic to a particular landing page.
This was all done clearly to reward better ads and landing pages for advertisers so that they click more and convert more - at the expense of a raw "per click" price. Aggressive alterations to this has been the course over the past 18 months with the idea of removing MFA and arbitrage - the real intent being that the advertisers that are left will be actual businesses etc. and so it becomes attractive for the small businesses and "real advertisers" to use Adwords.
The reality is that the group this system alienates is exactly the ones they wanted - no experimental advertiser gets past first base when using adwords because of high minimum bids, after all how can someone who has paid someone to develop their website be in control of the landing pages? It all becomes impossible to make a ROI.
Google didn't care about this as 80% of their income would generally be created from big agencies and this is very clear from the threads here - how many individual advertisers can even find out why they have to pay $5 per click. So they don't - they don't bother, and you can tell, the fill rate of Google ads is decreasing - the amount of shopping arb websites for keywords is increasing and Google is losing ad fill rates. This is why Google needs the Yahoo deal - Yahoo actually gets more money for key high paying terms per click than Google.
This new system is not an incremental development - it is a way to get more ads active in a new system and get people who know what they are doing to spend more. Otherwise why change it so much?
It is a new system - it is the system a company like Google needs when major banks go bankrupt, it needs income and it needs to ditch some of the standards that it had when the money was rolling in.
And it needs to do it without drawing attention to the fact it is making all search terms active - so calling it "first page bid" rather than "inactive for search" and "minimum bid" it is allowing itself to accept ads that it previously did not - not necessarily loads of them, just some of them - and maybe it can do it without it being seen as lowering quality standards.
The "trial" was open-ended but with the financial environment changing radically today it seems ironic that Google is suddenly ready to put it live.
Not meant to be cynical - just financial.
| 9:47 pm on Sep 15, 2008 (gmt 0)|
I'm finally seeing the "Active - Bid is below first page bid estimate of $1.25" thingie rather than a minimum bid.
However, this is showing for a keyword which only has 4 advertisers, so even at my bid of 50 cents, I AM on the first page.
And they thought they could trick me into upping my bid - heehee.
| 9:55 pm on Sep 15, 2008 (gmt 0)|
Wait a minute... so if there is no more min bid, and a keyword has only 4 advertisers, in theory I should have no problem sitting in that number 4 spot on the first page with a bid of 1 cent? Even if this keyword is marked as 'poor'?
| 9:59 pm on Sep 15, 2008 (gmt 0)|
I think what also I am interested to know in this new system is this:
If the new QS of keywords & ads etc. is determined at query time - which components of the adrank/qs are calculated because in my mind the existing elements that cause problems for people will not be re-evaluated.
What I am saying is - your landing page is not re-evaluated at query time in this new system, so the only thing that could be is the query term in relation to the ad content. So this means that landing page becomes irrelevant in large portions of the new system. How can you factor in the landing page quality score into a dynamic query system as the landing page has not been compared to the search term at point of query - unless adbot comes every time someone searches on Google!
Of course that is nonsense, so what I am saying is how can the search term relevance to the ad be determined at search time - well it can't, it can only be compared to the term you are bidding on, which (unless I am mistaken) is the existing system that we have all been using for the past few years.
So basically that is why landing page is surely less relevant in this new system - and why that would be the solution to the high minimum bid problem.
| 10:03 pm on Sep 15, 2008 (gmt 0)|
koncept, no as I understand it a keyword will still have a minimum bid and your ad and landing page will have a quality score that will determine what your minimum bid will be to enter the auction.
That is what Google has said - however, I reckon that is PR as the intention is to fill those ad slots as much as possible - you might not get away with a 1 cent bid to enter but you might get a minimum bid of something around what the others are paying.
Even if you would be seeing $1 minimums in the old system.
| 10:14 pm on Sep 15, 2008 (gmt 0)|
if there is still a min bid how am I to know what it is?
I'm doing a little experiment. on a keyword that used to have a min bid of 0.50 I have entered a keyword bid of 1 cent. Now it says "Bid is below first page bid estimate of $0.50" but I know there are only 4 advertisers. Now I'll wait to see if it shows.
| 10:24 pm on Sep 15, 2008 (gmt 0)|
that was quick,... the ad for that keyword is a goner.
So the question still stands... there are still minimum bids required in some cases, and without knowing what our minimum bids are (as this is no longer being displayed) how am I to know what my min is? This could cause confusion when trying to discover whether a keywords isn't getting impressions. (not enough searches or bid too low?)
The 'first page price' is not going to help here because it is inacurate in some cases, as in the one I describe 5 posts ago in which my ad is showing just fine on the first page for 50 cents, (it's old min bid), due to a low number of advertisers for the term.
| 10:25 pm on Sep 15, 2008 (gmt 0)|
And there is the issue that makes Google more money!
You don't know what bid will show your ad - you can try the 1 cent, if you don't get any impressions you would be tempted to go to 0.50 to get on the first page.
In telling you the bid to get on first page it is better than saying "minimum bid" as when you think of minimum 0.50 seems really high, but 0.50 first page sounds great.
In reality the system is saying bid 0.50 and you will have a chance of the first page - any bid below may or may not appear, it will depend on Google's system that will determine whether they will make money if you bid less than that.
Basically they have worked out that they will make 0.50 per click (or more accurately per #*$! impressions of that keyword) from the existing advertisers - remember the old system means that those existing advertisers are probably paying quite a lot as they too will have pushed the bids higher than the minimum bid.
| 10:31 pm on Sep 15, 2008 (gmt 0)|
What I tried to explain - theoretical explanation:
In the old system 0.50 was the minimum bid for example, 4 advertisers are willing to pay that sort of bid - in the new system in this case it is not in their interests to allow backfill ads so far short of the top range bids, i.e. you are going to have to bid near to them.
That is extremely simplistic as it is more likely based on lower bids and QS landing pages etc. - but the problem remains, in this case it is not cost effective for Google to let you enter the auction too far off what it thinks makes them the most money.
| 10:34 pm on Sep 15, 2008 (gmt 0)|
Also, just a quick heads up - ads that don't show on Google will show up on the search network sometimes, there is a different algo - the fill rate is higher on some partner networks.
I also think the main intention is to dynamically fill the search terms where there is a closer match between your bid and the bid landscape of the existing advertisers - so where the existing bids are 10 cents and there are only a few advertisers and you bid 10 cents (where you maybe were required to bid 0.50) then your ad will now show - and this is more likely on keywords that are broad matches of your key term, i.e. filling the long-tail.
[edited by: Swanson at 11:05 pm (utc) on Sep. 15, 2008]
| 12:24 am on Sep 16, 2008 (gmt 0)|
The whole black box is broken. I bid on my own brand/site name, mention the brand/site name in the headline of my ad, and the destination URL of course contains my brand/site name. Guess my quality score? Poor. If they can't get that right, the whole thing is just broken.
| 9:52 am on Sep 16, 2008 (gmt 0)|
I think Google should tell the top 3 page position bids, rather than just the first page. Not many will increase the bid to the first page requirement.
| 10:05 am on Sep 16, 2008 (gmt 0)|
I've noticed that if there aren't enough ads to make up the first page then min bid to make first page = old min bid, so there is no change there. If you are bidding on a niche which doesn't have much competition, whatever min bid is displayed, is probably what you need to bid over, to appear on the search results.
The bad part, is when there's lots of advertisers, and you don't have a bloomin clue, what bid you need to have, in order to get impressions.
It seems the 'start slow' method will work best. Start with lowish bids and if you notice the ad isn't being shown, then slowly up it.
Otherwise, everyone will be bidding top whack to get to the first page and this will mean higher CPC for everybody.
| 12:40 pm on Sep 16, 2008 (gmt 0)|
I bid on a high volume generic term for content as the minimum bid on search was too high for my budget.
Theoretically let's say the minimum for search is 30¢. I have always had my bid for search set at 25¢ & got no impressions.
Normal impressions on page 1 would be 10,000+ & since my ad is directed to a small percentage of this traffic I could never keep my CTR high enough even though the conversion rate on those that clicked through was high.
I am now getting about 1800 impressions per day with a good CTR at an average price around 18¢. I can only assume that I am now showing up on page 2 searches.
As a side note there is very little competition under this search term - only 2 or 3 advertisers.
| 1:03 pm on Sep 16, 2008 (gmt 0)|
So you're saying you can appear on the 2nd page (not appear on 1st), even if there's only 2 or 3 other advertisers?
| 1:57 pm on Sep 16, 2008 (gmt 0)|
Hey, are they going to show these scores in the AdWords Editor at some point?
So far the highest score I've found in 9/10, although I only looked through a couple of ad groups so far and I have hundreds to go through.
Wonder if it's like pagerank, where 10/10 is pretty much impossible, ork ork.
| 2:01 pm on Sep 16, 2008 (gmt 0)|
I have lots of 10/10 ... these seem to be the keywords that had min. bid of 0.01.
| 2:04 pm on Sep 16, 2008 (gmt 0)|
Yep, I found some.
I think this is actually pretty cool. I had some keywords that were hovering between ok and poor, and I had an idea what was wrong with them, but they were of lesser priority so were at the bottom of the list. Went to look at them now, and it told me exactly what I needed to do - which is what I suspected - loading time fine, landing page fine, but the ad text wasn't quite relevant enough to the keyword. They're more specifically-themed keywords in a general-themed ad group, so now I know I need to move them into their own group, write ads specifically for them, and they should be fine.
Now, if they just can get all this into the AdWords Editor, that would be the shiznit.
please pass that on, AWA - s h i z n i t (ork ork)
| 2:41 pm on Sep 16, 2008 (gmt 0)|
I '1 to 1' my keywords anyway, so think I might be ok.
I just hope the QS is included in the reports, so you can see at a glance what the keyword's QS is, without the need to go and check each keyword individually.
| 3:15 pm on Sep 16, 2008 (gmt 0)|
Despite the fact that one of my best relevant, converting keywords since 2003 has a first page bid 40% higher than my current bid, this information is great on the whole.
I see lots of 9/10 and some 7/10. But don't understand why it doesn't suggest anything to get 10/10. Just says no problems found.
I am seeing a pattern where my mispelling keywords have a first page bid higher than my current bid. I'm wondering if Google is not seeing the mispelling on the landing page and that is where I'm getting penalized. If this is the case, I would hope that this would eventually be addressed as I can't be expected to put mispellings on my landing page just to match my keyword. Anybody else seeing this?
Quick question for anyone. I ask myself this often in these discussions. How many Adwords accounts are there open these days? Last I remember a number of years ago was 200,000.
| 12:06 am on Oct 12, 2008 (gmt 0)|
this sounds like a long hard ride
| 3:30 pm on Jan 9, 2009 (gmt 0)|
A bit delayed, but for anyone that is curious the NYT just supplied the answer of my last post...
My main question..if you opened an account in 2002 and have optimized it ever since, and most of your ppc competitors opened their accounts in 2006/2007, what kind of advantage does one have over other advertisers when it comes to Account History and Quality Score?
| 4:32 pm on Jan 9, 2009 (gmt 0)|
Hmmm? I didn't see any info about advantage to long term advertisers in that article.
There was a comment from a reader about it at the bottom, but they were also just asking the question. I don't see a definitive answer. Just their speculation.
| 5:13 pm on Jan 9, 2009 (gmt 0)|
I know its not addressed in the article. Its part of the discussion since account history/longevity is a factor in quality score, which in turn effects cpc and position. Depending on how heavily weighted this is, the 89,000 advertisiers that opened accounts really early on would seem to have a significant advantage in QS over the million advertisers that started years after them.
A couple of years back, Google, (according to a rep anyway) was tampering with this "position normalization" idea by, seemingly randomly, swapping out newer ads into the top spots to counter this favortism towards companies that just happened to start in 2002. This was the explanation given to me when inquiring about peculiar activity when micromanaging a handful of keywords in an account spending nearly 1m/month.
I don't know what ever came of it, and am really curious to know how much of a factor this is or if this is still a concern.
AdwordsAdvisor...anything on this? Or was I given bad info...
| 5:22 pm on Jan 9, 2009 (gmt 0)|
I have no proof to back it up, but my spider sense tells me that 2 years is sufficient to establish a strong account history provided that you have decent quality in other areas. Beyond that, I doubt you have much of an advantage.
2 months on the other hand...
| 12:03 am on Apr 23, 2009 (gmt 0)|
I've been searching to no avail on this one:
Has anyone figured out any quantifiable differences between the new quality score numbers and their effect on cpc or position?
For example, let's say I've got a keyword that's costing me $5 for position 5 and I've got a quality score of 7 or 8. Then I do work to make it a 9 or a 10 (eg. micro target the ad and the landing page) how much of a difference to position or cpc will that make?
One simple theory I've seen is that the percent difference in Quality score may provide an equivalent percent difference in effective cpc:
For example, QS9 is 29% more than QS7 (9/7 = 1.29); therefore a $5 bid with QS7 would be equivalent to a $6.45 bid with QS9.
Another theory is that QS is like PageRank and is a logarithmic scale. If that's the case than the $5 bid would turn into, I don't know, say $100 ? That doesn't quite seem right.
Has anyone figured any empirical evidence or have any better theories?
| This 59 message thread spans 2 pages: < < 59 ( 1  ) |