My hand is UP
Shak
Just when you think you have it cracked eh.
With my volumes being so small though, that's unlikely to mean a lot to anyone other than me..;-)
Syzygy
Hands up if you think a serious ALGO change to Adwords has taken place over the last 2 weeks or so. My hand is UP
What makes you think that, Shak? Please elucidate...
Syzygy
Please elucidate...
Please do. I am confused about this as well. If* you are bidding only with US ads, then $ value is irrelevant. The basic algo is just CTR * Bid. I suppose they could weight one or the other but why? Other than this, the only thing that can really change is when your ad 'goes north', or how new ads get treated with initial rank.
But maybe you are talking about broad matching or something? I did have many more impressions yesterday but this was expected as the first of a string of holiday mondays, and you mentioned a time frame of two weeks. Why can't you elaborate on this. The post just invokes speculation.
not so easy to come steaming in anymore.
?
speak up chaps, instead of hounding me on MSN over this, and now I start the thread, u all go quiet.
google wont come looking you know. we DO live in a free society
Shak
The minimum clickthrough rate (CTR) of 0.5% won't be changing. However, when you submit a keyword, we'll now predict its CTR based on data such as the performance of your account and other accounts with the same or similar keywords. This ensures that we make our prediction based on a careful review of all available keyword data. Depending on a keyword's predicted CTR, the keyword may enter one of several states.
Shak
Or maybe they were just fedup with all the status questions on support :)
The most noticeable change you'll see within your account will be the keyword status labels for some of your keywords:
Previously strong, moderate, and at risk keywords will now be labeled as normal.
Keywords may also be labeled as in trial or on hold. *
Only in trial keywords may be slowed.
On hold keywords will not show ads unless they move to in trial.
Yep google have been making some huge changes over the last few weeks, especially when it comes to creating new campaigns and ads... If you havnt made changes or dont write new campaigns that often then you would probably never have noticed.
Taken from Google
We've streamlined the keyword review process so that your keywords and their related ads get the chance to run with less of a restriction on their accruable impressions.
The minimum clickthrough rate (CTR) of 0.5% won't be changing. However, when you submit a keyword, we'll now predict its CTR based on data such as the performance of your account and other accounts with the same or similar keywords. This ensures that we make our prediction based on a careful review of all available keyword data. Depending on a keyword's predicted CTR, the keyword may enter one of several states.
What this means in essence is it may be tougher or more expensive to start new campaigns if your old ones havnt performed well, or if you are opening a new adwords accounts.
I beleive the new algo was introduced about 2 weeks ago..Although they only confirmed it yesterday with the
New! Smarter keyword evaluation.
Our new performance monitoring system gives your keywords more opportunities to succeed. Learn more.
Is It good or bad?
Well that depends! if you are new to the game then its a terrible thing as google cant compare similar keywords or campaigns and hence this will mean cost will be a serious % more than the competition already in position.
If you are established then dont worry! just make sure you keep your entire account at a high ctr
In a way this new system seems to favour current advertisers and punish new people to the industry..
regards
Jamie
Additionally it sets parameters in terms of how your new kw's should be performing, based on the performance of others using the same kw's and relative to cpc.
...I think.
Syzygy
1. From: [adwords.google.com...]
Accounts that have proven to perform very well will be rewarded with a greater number of in trial keywords. The limitation placed on the number of keywords that may be in trial helps ensure that users continue to see the most targeted, relevant ads via the AdWords program.
What is this based on? If an account has 500k keywords, is it given more 'on trial' keywords than an account with 100 keywords? Or is it solely based on the overall CTR of the account? Or, even more obscure, is it based on the G only CTR of the account?
2. From: [adwords.google.com...]
borderline keywords slightly below our minimum clickthrough rate (CTR) of 0.5% may accrue at least 1000 impressions or more, depending on their actual performance. However, keywords well below the minimum threshold that are likely to perform poorly will be disabled sooner so that you can focus your budget on keywords that get results
Does this mean, that if historically, other's couldn't make the keyword work, then you can be disabled after only 100 or so impressions if you get off to a slow start? I've seen many keywords have a 0.25% CTR in their first 200 impressions before finally taking off and eventually those very broad matched keywords can end up getting a 1.5-5% CTR. Seems this system eliminates that possibility.
3. What I've noticed in the past week, is that there have been many occasions when a search should be bringing up the phrase or exact match ad/keyword - and instead it's showing the broad match version.
If 20 other advertisers are mostly using broadmatch, and doing ok with it, but not many are using phrase at all, and you come in and bid with all 3 variations, it seems that G's system would assume that the broad match will do better based on the other advertisers, and possibly show that keyword more.
If the other 20 advertisers really aren't showing broadmatch, but phrase match ads, does the new system take into account that it's actually the phrase match that's working? REF: [webmasterworld.com...]
4. A very common question is, 'Is my account just too bad to revive it, and should I just open a new one?'. In the past, it's always been viable to save an account. If an account now has a poor past history, and comes to an agency for help, is that account really savable now since there is a much more uphill battle as it's past performance will slow down 'on trial' and other type keywords?
5. I read this yesterday, but can't find it today. Basically, there was a paragraph on Google that read to the effect:
On trial keywords will show as inventory space becomes available.
Is inventory space considered Google search only for this, or on all properties? As the only CTR that matters is on G's site, having the keyword appear a lot on AOL, but not often on G, really won't help the KW to prove itself.
*bump*
AWA?
Now that the forum is returning to normal posting speed...
*bump*AWA?
Hmmmm. Well the Forum may be at normal speed, but not my tasks over here!
I'm running many hours behind on a couple of important tasks (one of them being the Feedback Report, which a very large number of folks are expecting in about four hours) - and I'm sort of scratching for time to post here on WebmasterWorld. eWhisper, I will get to your questions, though, ASAP. ;)
AWA
No worries AWA. Just keeping it on the radar screen. Two big changes in a few days must keep you hopping.
Heheh - thanks for understanding eWhisper - and you are correct!
Just saw your post as I was scanning for compelling stuff for the Advertiser Feedback report. ;)
AWA
Accounts that have proven to perform very well will be rewarded with a greater number of in trial keywords....
What is this based on? If an account has 500k keywords, is it given more 'on trial' keywords than an account with 100 keywords?...
Essentially, the higher the percenatage of keywords that one has which are performing well in 'normal' status, the more 'on trial' spots will be available. This is a case, though, where the exact details of the algo are not something those of us in support will know. As I learn more details, I'll certainly pass them on. ;)
borderline keywords slightly below our minimum clickthrough rate (CTR) of 0.5% may accrue at least 1000 impressions or more, depending on their actual performance. However, keywords well below the minimum threshold that are likely to perform poorly will be disabled sooner...
Does this mean, that if historically, other's couldn't make the keyword work, then you can be disabled after only 100 or so impressions if you get off to a slow start?...
What it amounts to is this: after watching every single impression and every single click on every single variation of every single keyword for that past 2+ years, we feel that we now have a pretty good handle on what will work. But in cases where we are not entirely sure, the keywords go to 'in trial' status - where they will run until it is certain whether they are meeting the standard or not. The number of impressions that it'll take to be entirely sure will vary - but I've pinged engineering, and am told that the number is in the many hundreds, at the minimum.
What I've noticed in the past week, is that there have been many occasions when a search should be bringing up the phrase or exact match ad/keyword - and instead it's showing the broad match versionIf 20 other advertisers are mostly using broadmatch, and doing ok with it, but not many are using phrase at all, and you come in and bid with all 3 variations, it seems that G's system would assume that the broad match will do better based on the other advertisers, and possibly show that keyword more.
If the other 20 advertisers really aren't showing broadmatch, but phrase match ads, does the new system take into account that it's actually the phrase match that's working? REF: [webmasterworld.com...]
I'm going to have to take some time to get my head around this - and will post again later on this one.
A very common question is, 'Is my account just too bad to revive it, and should I just open a new one?'. In the past, it's always been viable to save an account. If an account now has a poor past history, and comes to an agency for help, is that account really savable...?Absolutely, yes. And you would do the same things you've always done - which, to simplify, boil down to reorganizing the account into targeted Ad Groups, getting rid of too-general and otherwise low-value keywords, replacing poorly written and/or untargeted ads with well written and carefully targeted ads, and so on. The principles of success are the same as they ever were. Remember, as you make these changes, most keywords that are well chosen for their Ad Group will likely start out as 'normal'.
I read this yesterday, but can't find it today. Basically, there was a paragraph on Google that read to the effect: On trial keywords will show as inventory space becomes available.Is inventory space considered Google search only for this, or on all properties? As the only CTR that matters is on G's site, having the keyword appear a lot on AOL, but not often on G, really won't help the KW to prove itself.
I don't recall this detail, and wonder if you might be thinking of the fact that 'on hold' keywords can move into 'in trial' status when spots become available there?
AWA
I mean, jeepers - that's was one of my best CTRs in the campaign - how much better should it be? Is my competitor getting a 93% CTR? It just seems... wrong.
I read this yesterday, but can't find it today. Basically, there was a paragraph on Google that read to the effect: On trial keywords will show as inventory space becomes available.
Is inventory space considered Google search only for this, or on all properties? As the only CTR that matters is on G's site, having the keyword appear a lot on AOL, but not often on G, really won't help the KW to prove itself.
Are you preferred to either one of these paragraphs?
from [adwords.google.com...]
in trial
There are a limited number of in trial keywords per account, so when this limit is reached, additional keywords will be moved to on hold status with priority given to those keywords with higher predicted CTRs.
on hold
As more space becomes available, these keywords will move into in trial status and start showing ads, with priority given to those keywords with higher predicted CTRs.
I believe everything about keyword status is based on Google search only.
To AWA,
The above two paragraphs are in contradiction. I believe "higher" in bold as in paragraph 1 should be lower.
Essentially, the higher the percenatage of keywords that one has which are performing well in 'normal' status, the more 'on trial' spots will be available.
Do the "normal" or "in trial" keywords in the paused campaigns or paused ad-groups include in this reward or limit?
I was surprised to see terms that have a 7% CTR being put on hold.
I mean, jeepers - that's was one of my best CTRs in the campaign - how much better should it be? Is my competitor getting a 93% CTR? It just seems... wrong.
Ditto!
For the layman that means we ony want the highest pay-out ads showing during the peak periods of November- March. If you increase your pay-outs inventory space will become available.
I have a fair amount of adgroups that are promoting some name brand products with various spellings of the name.
I am wondering if - and precisely how much - keywords that get low impressions will harm my account?
Right now my overall account CTR is at 0.8%. It is only that low because of some content sites - and the inability to, on an ad-group basis, decline inclusion in content match. My search CTR is higher - I cannot easily give a number unfortunately - but at least above 1%, maybe above 2% - hard to say. Could be as high as 3-4% - unfortunately when an ad gets 1000 content impressions suddenly due to a popular website having one of my keywords.... Well nothing you can do except move the ad, pause keywords, or pause the ad.
Overall - my keywords are well targeted - and I would occasionally have a keyword disabled in the past due to high volume of searches (seasonally for instance) that didn't result in clicks. But overall due to the targetting of my ads - I have decent CTR. (Sometimes less than 0.5% - but still active due to less than 300-500 impressions)
With the new keyword um "scheme", I have noticed many keywords go to disabled. One ad that has been running for months has a keyword disabled that only had one impression. (And no clicks) Another ad has an search CTR of 1.7% - with my most active keywords at 4.1%, 2.2%, 0.8%, 0.9%, 2.7%. A few more with 0%, but still active due to low impressions... And then a bunch of keywords (6-8) with no impressions that have all been disabled.
So am I penalized only for low CTR - or does impression count factor into it too? Because if it does - I now am going to have to be a heck of a lot choosier instead of capitalizing on decent CTR rates on rarely typed keywords.
I'm concerned. I make money off leveraging clicks off some very rarely searched keywords - and hence I have some obscure keywords. If I have to narrow my keywords to keep them ALL highly targetted. Well... owch. I'm adding a handful of products on a weekly basis - and am greatly concerned about this keyword situation as my existing keywords could ruin my ability to promote these new products?
I guess to summarize - let's assume I have:
o 100 Ad groups
o 10,000 keywords
o Overall 2.5% search CTR
o 200 keywords in the 3-15% CTR range
o 200 keywords in the 0.5-3% range
o 100 keywords in the < 0.5% range
o 500 keywords with less than 10 lifetime impressions
Is my account at risk for not being able to add new keywords or new ads?
Thanks for any thoughts you can give :)
