New! Get more clicks automatically with the Google Budget Optimizer™ tool. Let the Budget Optimizer actively adjust your keyword Max CPCs to yield the highest possible number of clicks within your target budget. Learn more
I'd love to check this out, but the link doesn't work. :(
Perhaps our competition has BO. (sorry could not resist).
Could this be the "calm before the storm" or is BO, in fact, and "Dampening filter" of a sort?
Or it could just be "noise" in our case.
We have been in a bidding war (ish) of late so perhaps they "blinked". I dunno.
But a 20% drop is significant, we are pretty happy about it. Hopefully it will hold.
I am just speculating that perhaps they have BO and that is the cause.
Strategically I can see why G would want to "spread the clicks" around to various "under-utlized" words to more fully spend the budget. It is like an automated "niche-finder" (hopefully).
This could perhaps have the intended result on taking some of the stress off the highly priced words and reducing the CPC as the $ is spread elsewhere.
Personally I am inclined to think it is a "fool's errand" because there are words we *know* can get better CTR but we also know that (for many reasons in our case) those words don't *convert*.
I'd really like to know if they tie-in conversion data over time. If so, I'd be inclined to try it on a portion of our account. (in a "quiet" ad group for sure)
If not, then I'lll pass.
I checked and our conversions are also up 20% the last 2 days.
We sell a high priced service so we track average page views per user as our "conversions" .
20% increase in average PV's for us is within in the realm of normal traffic. More so for Monday than a Tuesday, but it is up.
Our average position has also increased.
SO for us, CPC down, Convsions Up, Position Up.
With a pretty competitive marketplace like ours I am <i>begining</i> to suspect something is up.
But I won't be convinced unless it holds for awhile.
Problems with your competitor? Send an anonymous tip to them, stating that using the BO is ultra-secret-prof advice at the moment...
I think I'll give that a try... :)
- Jan
(Would be nice for normal AdWords as well)
BO strange behavior: my daily costs aren't anywhere near my daily budget, but still the BO sets some max CPC's at 5 cents on positions 5 through 12 (!) Even with high CTR's. That doesn't sound like maximum possible number of clicks.
It seems like its positioning the campaign for a 'spike' in interest, and then it puts your campaign
up there in an optimum way. I am not certain on my performance evaluation for it at this time.
I do know the last few weeks (before optimizer came out) I have increased my daily budget as they recommended, and have not regained the kind of results I was seeing like when I first began the campaign. It also didnt help they removed my ad for (2) days on their whim, and it took several emails to get it back running....then I had to make a new campaign up because that one no longer was able to be broadcast after they messed it up.
I've seen the page before, but can't remember how I got there....
Thanks
Thinking of switching it off.
Can someone tell me how to get to the page that shows what the other ads are paying per click?I've seen the page before, but can't remember how I got there....
NO. This kind of info is not avaliable for AdWords. All we have you can learn here (mainly point 4):
[services.google.com...]
But I think, that this is only a part of the story, because AWA posted here [webmasterworld.com...]
this info:
The short story is that, yes, the ads positioning algo has in fact shifted from considering only keyword ctr, and now folds-in information about the relevance of the ad itself, including the Display URL.
If I undersatand it well, this formula is not true anymore: Rank Number = maximum CPC x CTR
The reality looks like this: Rank Number = maximum CPC x CTR x RAD
where RAD is a variable that depends on relevance of the ad.
BUT I HAVE NEVER SEEN this formula ofically, it is how I understand the old info + AWA info.
The reality looks like this: Rank Number = maximum CPC x CTR x RAD
Well that is NEWS to me. When did that change? How come G doesn't email advertisers with sucha fundamental change. For weeks I have been banging my head to understand why Ads with much higher CTR also have much higher CPC on average. Weeks! They must not have enough RAD.
So how would RAD be measured? A separate CTR of the ad by keyword?
Any ideas?
We have been an adwords advertiser since the begining and I must say (AWA if you are listening) that I have always been a pretty big G booster.
In terms of our (my) relationship with Google as an advertiser it doesn't seem like we are in a kind of partnership much anymore, but rather Google is going to some lengths to hide important information from us.
We spend tens of thousands per year, and I'm sorry, but I think we are worth an email outlining such a big change as this.
First, it wants to optimize three results (CTR, max CPC, and conversions) with only one variable that it can control (max CPC). This is, in mathematical terms, an underconstrained problem, which has no single solution. They might treat the max CPC of each keyword as a separate variable -- but the number of results then grows too (same results by keyword). Actually it would be very very helpful if Google would just say what the controlled variables and measured results are -- not the algorithm, which I'm sure they consider proprietary, but the results and controlled variables. That's just basic information for their customers.
Second, as pointed out several times, the result the advertiser really cares about is conversions, and they only have that information for advertisers using conversion tracking. And even for the latter, Google does not know the value of the conversions.
Third, there are many human factors involved, not only things like the probability of clicking based on position (which can be predicted statistically with good accuracy), but how that interacts with keywords (hard to predict), the effect of today's headline news (very hard to predict), etc.
Over the years I've been involved in situations involving underconstrained systems, unmeasurable outcomes, and/or human behavior. The results have generally been less than satisfactory.
Edward
The reality: my clicks went from around 500 down to 1 on Saturday.
BO gives the following:
Over the last 30 days I spent around $390 for about 7800 clicks
BO suggests I spend $450 over the next 30 days for about 7850 clicks
Or spend $635 for 9600 clicks
So, spend 15% more for less than a 1% increase in clicks, or spend 88% more for 25% more clicks.
That doesn't strike me as a particularly good deal!
Is anyone doing a lot of ads at the 5c level and getting better advice from BO?
is it still in beta? or is this the final code for the bo?
Rank Number = maximum CPC x CTR x RAD
Sorry for the diversion from the topic, I think that Google also doesn't use the straight CTR but rather the CTR weighted by the position of the Ad. What I mean by this is that 1% CTR in the premium position is given less weight when compared to 1% on the first position on the right hand side which might be weighed less than a 1% CTR for the second positon on the right hand side and so on. This is the only way I am able to explain, how I have been able to climb up into the premium position for many of the words despite a huge gap in CTR between the Premium Position and the right hand side. Without this weightage it would be very tough to unseat an existing competitor from that premium position.
What they told me was at odds with what AWA said in the other thread, so I don't know exactly what is going on.
The reality looks like this: Rank Number = maximum CPC x CTR x RADwhere RAD is a variable that depends on relevance of the ad.
I called Google today about this RAD business and they emphatically and categorically deny that there is anything which determines rank other than CPCxCTR.
This sort of illustrates the problems inherent in publishing a speculative and unverifiable "formula" in a public forum, as if it were a fact.
What they told me was at odds with what AWA said in the other thread, so I don't know exactly what is going on.
Not really at odds, nyet. It is still Max CPC x CTR. The difference is that now, the CTR of the ad copy itself is factored in, instead of it being solely the CTR of the keyword. Which only makes sense, IMO, given that it is the quality of the keyword and the particular ad it brings up that defines relevance, for a given search.
And I'd like to emphasize that the key to success is the same as it ever was - which is to show an ad that is highly related to the keyword that made it appear.
BTW, as I also mentioned in my previous posts about this, many weeks ago, this incremental change to the algo is one part of an ongoing effort, first mentioned here, in June of 04:
[adwords.google.com...]
Quoting from that page (emphasis mine):
...In the next few days, we'll introduce the first in a series of ad quality improvements designed to provide the most relevant, targeted ads possible...
Lastly, I'm responding to this post at the risk of furthering what seems to be the 'hijacking' of this thread, which really was intended to speak to another topic altogether. My apology, but it seemed this question deserved an answer.
AWA
Is anybody still using the Google Budget Optimizer?
Regards,
Yes, I just started using it on three *very low* budget campaigns that I want to fire-and-forget (at least until I get controls to run ads at particular times/places and feedback through the console about when and where (geographically) they do well).
I'm hoping the optimiser knows some things I don't, ie has access to some variables that G will not show us, such as some of the above.
Rgds
Damon
