| 6:34 pm on Jan 31, 2003 (gmt 0)|
>"raise your bids and we will leave you alone?"
It won't matter if you raise your bids, its because not enough people are clicking on your ads. You need to focus your keywords better and/or improve your ad copy.
| 4:22 pm on Feb 1, 2003 (gmt 0)|
Low CTR means that not enough people are clicking on it. I think the lower limit set by Google is 1%.
That usually happens when the keywords are too generic. We have some specialized product specific keywords that hit as high as 17%, average is around 2.5%.
| 9:24 pm on Feb 1, 2003 (gmt 0)|
I have been learning how dumb I am. What I feel is targeted seems to be constantly rejected or I am told they are Gold Yellow **
I edit and edit. I gotta tell you, I do not see how you can say proce has nothing to do with it. If you are willing to pay higher price, you get higher placement, more visibility, more click throughs = more money spent. I disagree that it is not about money. Please do not take offense at my forwardness. I am looking for your expert opinion. Although I am very impressed with the work and seemingly personal messages I have received from AdWords, I still am overwhelmed at understanding their message. Too smart for me. I need simple directions. Do this to equal this.....
| 3:57 am on Feb 2, 2003 (gmt 0)|
" do not see how you can say proce has nothing to do with it. If you are willing to pay higher price, you get higher placement.."
We use about 300 keywords. for 200+ of them we pay 5 cents. the highest we pay is 38 cents.
| 5:09 am on Feb 2, 2003 (gmt 0)|
i assume you are using the following targeting strategies which should improve CTR
1. negative keywords
2. targeting by countries or languages
3. quotation or  keywords
4. specific ad copy for specific keywords
| 6:13 am on Feb 2, 2003 (gmt 0)|
I just tried the Great Google Ad word debacle and failed miserabley. When I tried to sell my blue widgets as blue widgets and was willing to pay the .35 per click I was told I wasn't getting the CTR. So I changed to key words that were getting the hits and used a good discription and title and still they rejected it. What I found by doing some research of my competitor was that he was willin to pay $5.oo per click so he remains #1 even though the search terms were the same as mine. Money talks!
| 6:50 am on Feb 2, 2003 (gmt 0)|
You don't have to be #1. In fact there is some evidence that the #2 position is actually better.
"$5.oo per click so he remains #1 even though the search terms were the same as mine..."
How much he pays for adwords has nothing at all to do with his ranking in a Google search.
| 6:57 am on Feb 2, 2003 (gmt 0)|
<<If you are willing to pay higher price, you get higher placement>>
This is true, but only up to a point. Even someone willing to pay several bucks a click will get ads or keywords cut if they don't make the minimum CTR.
There's nothing for it but to keep tweaking. Be sure you understand Chiyo's advice, especially about negative keywords.
| 8:00 am on Feb 2, 2003 (gmt 0)|
Absolutely. Welcome to the world of PPC, or should I say web business of this century. Free or cheap advertising on the web for commercial sites is declining fast, and was only possible while Search engines developed and tested their algos for the past 5 or so years.
All PPC or PPI do some form of checking for relevance to make sure that browsers will get useful results, but after that, they of course will display the ads that provide them the highest revenue. That only makes business sense.
But given that, Adwords also has a built in relevancy check in that they wont continue to display keyword combinations that have a CTR of less than .5% over 1,000 impressions. That is a good way of ensuring relevance for their end user customers.
If you are not using brackets, some of your keywords will prompt an impression and some wont, depending on what other words other people are using.
For example with keywords of ¦white elephant¦ you may get impressions when people type in ¦white elephant trunks¦ but not ¦white elephant sale¦.
¦sale¦ may be an expensive word, while ¦trunks¦ may be a cheap word.
That is why increasing your max bid for a campaign increases impressions for these "wildcard" type queries, (in this example, those that also include the word (sales)) as you start to pick up higher value queries. You will also be in different orders in the Adword columns depending on others' keyword combinations and max bids.
So max bidding is not the same as say max bids in overture, and understanding these differences are important in those cases. Of course using [white elephant] you will only be competing against others who specify, like you, that you only want impressions for people who type in ONLY those words, in that order, and no other terms. In that case Adwords is more similar to OV I think.
I would guess that there are some areas in which Adwords is not the best method of advertising (for example costs are very high for several of our areas compared to OV, and we just nix 'em) and you should look at other options such as Google premium/sponsored links, OV, Findwhat, espotting, banner exchanges, affiliate programs, and advertising on content sites.
I dont really think Adwords will work for everyone, in every area. Best areas from a very quick experience, are areas that are not too market-niched and not too general (somewhere in between) and have good synergy with Google's demographic profile.
Remember to use plurals, diff spellings, and different terminology for the same thing (adverbs, nouns, verbs etc.) and see which terms get the best CTR and nix the rest.
While you are testing a campaign for relevance and you expect to get many impressions and cant check every few hours or so, remember to give Adwords a max daily spend, which will extend the time for your words to start getting starred or whole campaigns ditched, and give you time to modify your campaigns in the meantime. That way you wont get ditched in less than a day and you can gradually "craft" a good CTR and "ROI" capmpaign. Nothing different than mainstream ad campaigns - running small sample ad tests before rolling out broadly is commonly practiced by ad agencies.
[edited by: chiyo at 8:10 am (utc) on Feb. 2, 2003]
| 8:04 am on Feb 2, 2003 (gmt 0)|
And also check out including commonly misspelled words.
For example, we sell inverters. So we also have keywords for "invertors". If you sell widgets, perhaps you would want to also use "wigets".
| 8:27 am on Feb 2, 2003 (gmt 0)|
Yes, and moreso... it's cheap money, but only if it converts.
If you have a $50 product that only took $5.00 and no sales help to convert that $45 net profit after marketing expenses.
Therefore your Adwords can't just be about keywords but very good matches to next page.
The real pain starts when 10 users click and only one purchase which is a net loss (break-even on marketing expenses but the product cost something to make.
I find the best Adword campaigns are short term promotionals.
Run two campaigns one maybe at your regular price (all products but a separate one at discount on a specific product (say the one you have the most inventory for) at 40% off.
Once the inventory of this product drops below another... switch, and so on.
Normal net returns works out at about 75% but the faster conversions get inventory out the door faster.
The faster it go the more you buy at a cheaper price which now increases that net return to 85%.
Still further if your shipping large volumes of product daily you now have leverage with shippers (once in a while users are non-profiable to them), which now increases that net return to 90%.
How many other consumables do you need to purchase to push product out the doors... buying in larger quantity means more savings and that would likely get you up to net 95%.
Darn close to regular price in my books.
| 11:11 am on Feb 2, 2003 (gmt 0)|
"that would likely get you up to net 95%..."
What possible product could you buy for $5 and sell for $100?
| 11:21 am on Feb 2, 2003 (gmt 0)|
Was using percentages and imaginery numbers... but actually CD-ROM/SilkScreen/Jewel Case/Shrinkwrap/Shipping Package $1.35
| 11:28 am on Feb 2, 2003 (gmt 0)|
Look at it from Google's point of view.
CompanyA has an ad that has a CTR of 5.0% and pays $0.10 per click.
CompanyB has an ad that has a CTR of 1.0%.
How much does company B have to pay to get above company A?
Well, google makes 50 cents from each 100 impressions on the company A ad.
So company B needs to generate google at least as much money, and for CTR of 1.0% that would be 50 cents per click.
Either way, google makes the same amount from each 100 searches.
That's a brilliant approach of an open market.
It's like an automated compromise system for multiple parties :)
| 11:29 am on Feb 2, 2003 (gmt 0)|
|What possible product could you buy for $5 and sell for $100? |
"How to become a millionaire in two weeks" e-books :)
| 12:48 pm on Feb 2, 2003 (gmt 0)|
Could someone direct me to post/thread where brackets are discussed? I would like to understand them better. Thanks to all of you.
3. quotation or  keywords
| 1:07 pm on Feb 2, 2003 (gmt 0)|
I agree with glenv about them using tricks to get you to spend more money. Except in my case, they did it differently. I bid 4p / click on the exact same ad copy that was getting me 21 - 25% CTR on Overture. The ad NEVER (literally never) showed up on Google over a 3 day period, despite the 4p being more than the minimum bid for that keyword / country. In order for the ad to show up at all, I had to raise my bid to 6p! And then it displayed at position 19 despite me knowing that at least two other ads above it are bidding less per click!
If it were a case of a steep learning curve then fine, that's a challenge and I do stuff like that all the time. But no stats, no progress reporting, sly methods of increasing click payment, no customer support except a pre-scripted "you are thick, try this... step 1, step 2..." email and certainly no results (except loss of my money) - what a joke! At least this experience explains how Overture thinks themselves justified in increasing their minimum PPC - they offer a working service.
| 4:11 pm on Feb 2, 2003 (gmt 0)|
"at least two other ads above it are bidding less per click!.."
How do you know they are spending less?
Unlike Overture, I can see no place where you can see what an individual bid is.
| 4:18 pm on Feb 2, 2003 (gmt 0)|
windsun is absolutely correct. There is no way to tell what your competitors are bidding, unless they are friends and have told you and have not lied!
It is not an auction type system like OV and several others, - it is substantially different.
If i was Google i would rank ads very simply..
CPC * CTR - and maybe give new ads a boost while the CTR stabilizes.
CPC * CTR =? also gives google the best ROI for them!
| 5:52 pm on Feb 2, 2003 (gmt 0)|
Because one of the ads was placed by someone I know and another ad I read about indirectly in a forum post, which may or may not have been true. What I do know is the bidding term is necessarily generic and the products retail at between £1.50 and £10. With that in mind, a lot of people must be getting exceptional conversion rates! In fact, if the ordering really is as Google specifies (see below) then I'd estimate the affiliates on page 1 needing a 25% conversion rate to break even!
The CPC * CTR is the way the ads are ranked, as clearly stated in the Google FAQ. My point (or rather, a sub-point of my main one) was that seeing as my ad went in with 0 CPC, how will it ever be ranked higher? Who will click on a zero-ranked ad? And how do I tell my performance when I have had no stats whatsoever from Google, since the day I started, 4 days ago?
| 7:36 pm on Feb 2, 2003 (gmt 0)|
GhostMonkey, I remeber reading somewhere that Google give you an average CTR (based on what your "neighbours" get) during the first 1000 or whatever impresisons, giving time for your ads to develop CTR data of their own.
| 8:08 pm on Feb 2, 2003 (gmt 0)|
"What I do know is the bidding term is necessarily generic.."
That right there is a major problem. Generic keywords from what we have tried are not very good. You may get a lot of clicks, but not many sales.
And you have to be careful about bidding too high on common words - it does not take long to get into a money losing situation.
This was brought up with a vengeance last Sept on Overture:
3 or 4 companies had all set their bids to auto on a relatively common keyword. Since it continued to ramp up the bids as another one went higher, the bids finally reached over $25 before someone figured out what was happening and pulled the plug. People were having a great time clicking on them.
Since they were all trial lawyers nobody felt sorry for them though.
| 9:09 pm on Feb 2, 2003 (gmt 0)|
Jonboy - yes, this is what I mean - my ad went straight in with 0, which isn't right. Perhaps I'll delete it and try again.
WindSun - I reckon some trial lawyers could still make money on $25/click! In actual fact the keyword probably isn't as generic as you think - it's not like "games" or "books" or something, just a bit more generic that maybe it *should* be. Trouble is, when I have a look at the Google suggestions for more specific phrases it always says the forecast clicks are <0.1 i.e. zero. It even says it for phrases that Overture says are searched on 1,000's times per month. I suppose I could ignore it...
| 2:18 am on Feb 3, 2003 (gmt 0)|
"Google suggestions for more specific phrases it always says the forecast clicks are <0.1 i.e. zero..."
<.1 is not zero though.
In fact probably 1/4 to 1/3 of our keywords say that. Some only get 3-4 hits a day - but on one the buy rate per click through is 27%, and the CTR is over 10%.
Google does not kick ads out because they have a low impression rate - it kicks them for low CTR. If you have a word that only gets used 10x a day, but gets clicked on once, it will be fine as far as Google is concerned.
Granted, it takes time to find all these obscure keys, but the upside is that we are only paying 5 cents for most of them.
| 12:40 pm on Feb 3, 2003 (gmt 0)|
Well, after deleting and recreating the ad (now it works fine) I selected a few keywords with <.1 estimated clicks and it looks like you're right. One got several clicks in the first few minutes, so it looks like it's best to ignore Google's "estimates" (it would be easier if the stats were working, I have to trawl through my logs at present).
| 8:14 pm on Feb 3, 2003 (gmt 0)|
We have one really obscure keyword (a manufacturers specific part number) that comes up maybe 4-7 times a week. But the CTR is over 50% and the buy rate from our Yahoo store is nearly 20%.
I think the ROI on that one must be around 5000% ;)