I wanted to know what I had to do to to get my Adwords up there, so I experimented. Here's what I found:
------------ CTR -- Bid ---- UK monthly Imp -- Bid x CTR
Keywd #1 -- 3.61 -- £0.58 -- 12,000 --------- 2.09
Keywd #2 -- 1.86 -- £0.97 -- 3000 ---------- 1.80
Keywd #3 -- 0.88 -- £2.08 -- 6000 ---------- 1.83
Note: All keywords were geo-targeted to UK only.
"Bid" here means the minimum I have to bid to see the ad sent north.
What's interesting here is that the number (Bid x CTR) is very close in all cases; indeed for Keyword #2 and #3 it is so close that it is probably within the margin of error caused by (a) that CTR may change during testing, (b) that reported CTR may not be up to date with real CTR, and (c) that Bid must be a whole pence so rounding is involved.
Maybe BidxCTR for Keywd#1 may be higher because it's has higher volume of impressions, maybe crossing a threshold of some kind. From what I read on WebmasterWorld pricing for Premium CPM ads displays "bands" like this, e.g. $10, $15, but not $13.45!
What's going on here? My guess is that it has some link to pricing for Premium ads. Probably Google wouldn't want it to be cheaper for people to get into Premium spots via the adwords back door than via the big committment front door they try to sell. You'll notice how if you multiply BidxCTR by 10 (to move from CTR as a percent - per 100- measure, to clicks per 1000), the numbers you get (£18, £20) look very like normal Premium CPM prices.
Why is any of this interesting? Well, it's not interesting if you have agressive competitors who have high bids; their high bids may force you to pay the high amount you bid to get north into the Premium spots, making it better for you to buy CPM the regular premium way.
But if you're like me and (a) have few and timid competitors who only make the minimum bid, and/or (b) are keen to stay at the top of the pile anyway, then you can afford to raise your bid and get into Premium land: you won't have to pay any more for it than usual. :)
If other WebmasterWorld Adwords posters could experiment with raising the bid price to find when they go north, and report the results in the format I have above, we might be ale to work out precisely the formula Google is using. ;) Particularly what is needed are some keywords with higher volumes.
Some of my data:
impr ctr max actual
167000 2.6 0.15 0.11
1000 5.3 0.15 0.09
200 7.0 0.15 0.09
I will share the results
A question...how can you determine if the "Up North" ads are promoted Adwords ads or CPM ads?. Surely if they're CPM ones they are guaranteed to be in that position, so you could increase your bid to $100 per click and still not "Go North".
In fact you'd spend a lot of money and probably "Go Under" :)
Basically, you gradually home in. You vary the bid in the Adowrds control panel, save the changes, wait 60 seconds, and run a google search on the keyword to see where you ad appears. (it helps to have 2 browser windows open - one in the adwords control panel, one in the google search page).
You work by find an "upper limit" bid where you know the Ad goes north, and a "lower limit" bid where it doesn't. You then test a value in the middle of these. Depending on whether the value sends the ad not or not, this middle value becomes the new upper or lower limit; then repeat the procedure.
An example should make this clearer.
Say you're currently bidding $.50 and you're #1 but not up north. Then $0.50 can be your lower limit.
Try raising the bid to $3; assuming this does send your ad north, you have an upper limit.
Now try a value in the middle of these 2 - in this case $1.75. Let's assume that this succeeds in sending you north, in which case it becomes your new upper limit.
Now you try £1.13, which is halfway between your current lower limit of $0.50 and your current upper limit of $1.75.
etc.etc.
Here's how a whole experiment might look. Yes means that bid sent you north, no means it didn't.
0.50 no
3.00 yes
1.75 yes
1.13 no
1.44 yes
1.28 yes
1.20 no
1.24 no
1.26 yes
1.25 no
Therefore in this example $1.26 is the threshold for going north (the minimum value needed to go north).
TIPS:
1) if your number of impressions on a keyword is very low, be patient and wait 60 seconds before searching. If you try every few seconds looking to see if your ad has gone north yet or not, you might significantly increase the impressions and thereby affect your CTR.
2) sometimes the threshhold is unstable by a cent or so; I ssupect this is because minor changes in the CTR during the course of the experiment.
3) You can save yourself time by selecting your original upper and lower limits carefully. Based on the results in my original post, it seems that the threshold is likely always to be between (2.5 divided by CTR) and (1.3 divided by CTR) times your CTR. Be warned though that the figures I inferred of 2.5 and 1.3 referred to UK £, so for the US try 1.95 and 3.75.
Example: if your CTR is 2% try starting your lower limit at (1.95 / 2 =) $0.98 and your upper limit at (3.75 / 2 =)$1.88; if my experience is representative, you should find the threshold is between these.
Ross: You're right, I don't suggest one try's this unless there a Premium slot free. If there is 1 slot free, and the other slot is taken by a regular premium customer, there doesn't seem to be any problem, the above procedure worked for me the same as normal. If the single guy in premium is already another promoted adword, I don't know what happens. Can you have 2 promoted adwords up there at the same time (assuming no regular premium customers)?
Like the methodology, the old binary chop eh?
fantastic bit of work.
only worry is will G now change the limit to go up north.
I wouldnt have thought so, if its CPC and CTR is higher up North, combined with NO premium listings, then why should they care.
Lesson #1, make sure you buy premium listings to be absolutely safe.
Lesson #2, attend this forum on a regular basis to get the latest class on Adwords, I really gotta start paying attention to Adwords, have gone too long with Premium listings.
Lesson #3, attend Boston, for the official Adwords class :)
(go on Brett, sort it out)
Shak
In a recent post in another thread (at the end of the long 79 message one about Premium)Algebrator reported that he'd found when comparing Adwords position #2 and premium, CTR had doubled while conversion fell by 33%.
One poster elsewhere said that Premium salespeople said that it got an average of 50% higher CTR.
Beware of that falling convwersion though!
did not know it was so easy to (buy) play high score in real life.
Will make interesting experiments on the change of click throughs and ROI.
CTR = 18.4%
CPC = $0.65
This has resulted in more clicks and an increase in sales. Thanks for the experimenting ideas. I will make good use of the ideas here.
Joe
Thank you
Let's say I am targeting only one keyword in an adgroup.
I am getting currently a 9.9% CTR . Average current CPC is .24 US ..
You are saying I can drive this into the sponsored spot?
When you guys say "go north" are you talking about the two main ads shaded in either pink or light blue that run full horizontal in results? They both say over to the right, "sponsored link". Does this method of back dooring it work in the US version of Google?
Up North is the same as Northern banner, Top Listings, Premium Listings (in other words the top 2 shaded results).
I am pretty sure it works at all of Googles locations, although have not tried it myself (honestly).
Shak
I went ahead and bid high enough to get in the premium placement (up north as you put it). I was successful.
I decided to leave my placememt there a couple hours to monitor ROI on new placement.
I came back a few hours later, really had not seen anything happening and pulled back my bids to their normal state. I had to do this since I would be away for the rest of the day and did not want to take the chance of a run away campaign.
Needless to say, the next day I noticed activity on about 5 of my 8 main ad groups had rolled over and died. This was strange since even before I experimented at making the bids "go north", the keywords were doing well with a range between 9-24% CTR.
I watched them all day today and still nothing. I decided then that I would go in and set back my biddiing a few pennies on each group to see what happened. Interstingly enough, they have now almost instantly started up again.
I am at a loss as to how or why the system would stop all bidding. Any one have any theories?
Great tip so I thought I'd try it out, but I had a strange problem.
The google page shows the top two places already taken. No1 is a major corporation (and I would expect to see them there) but number 2 is a much smaller competitor of mine.
To see if they had bitten the bullet and started spending some real cash I upped my bid to $4.00. Sure enough their premium placement disapeared but I didn't replace them. I just sat above them on the right hand column.
I've played around with the bids and found that if a bid a more reasonable $1.40 then I stay above them and stop them getting the 2nd premium place but neither I or anyone else move into it.
If I drop the bid even a couple of cents then they jump back into the number 2 spot.
What's it going to take to get me that spot?
Peter
I notice that in adwords the title is limited but many premium sponsors I see have much longer titles.
It sounds strange though. I can't see what the point of that is for Google. Has anyone else found this to be true?
VisitThailand:
If you're naturally promoted from AdWords to premium the your titles and decriptions are identical, you have no way to change them.
I have heard though (I think) that naturally promoted listings (can?) have a redundant extra space in the middle of the description, where the 2 lines of the Adwords dexcirption are glued together to make a single lined premium description. Don't know if this is true.
[google.com...]
Line 1/title: You are allowed to enter up to 72 characters, including spaces.
Line 2: You are allowed up to 80 characters INCLUDING Spaces and Visible URL.
hope that helps.
Shak
(Google gonna be loving this thread)
As was noted earlier in the thread, CTR x Bid x 10 = CPM figure. I can't prove thats what happens, but as the Premium slots are CPM, surely the qualification criterion for them should be also?
Now I've just got to keep an eye out for falling conversions ;)
Peter
In a highly competitive environment, if you want a premium spot, buy a premium campaign. Do not use the method described in this thread. Why? Because your competitors will do the same and you will outbid each other until you pay 10.00 dollars per click or more.
In less competitive areas, this method is great and you can improve your ROI greatly.
This 1.8/2.0 ratio makes it almost impossible to make it stay up there for more then just a few searches. Even by doing my own research i modified the total hits and made it fall out of that ratio.
Logically, since Premium drives more hits and clicks, the ratio changes and you ad falls back again to the Adwords level. If you target rare kws then just 1 click can radically change your ratio so you would need to adjust the CPC value every 5 minutes or so to maintain your premium spot!
Anyway maybe you guys can tell me how you do it because i don't have a clue right not on now on to go up there but rather on how to "STAY" up there.
Thanks,
Mtlinfo
If your keyword is rare hopefully you have no competitors, so you could safely (with []exacvt match) put your bid ludicrously high at first to ensure your ad stayed up north until its CTR had become stable.
But waht exactly is the most important thing for Google to leave your ad "up there", having a CTR X Bid ratio of 1.8/2.0 like you said or a just a very high CTR?
Mtlinfo
