Using the three key words independantly will trigger fewer results, and will not neccessarily be reflective of your market. If, for example, you have 'widget' on a broad match you may end up showing in response to queries such as "beer can widgets". Not ideal as you only sell 'Blue Widgets' not beer can widgets.
To open up your opportunities, ensure that you are on broad match and use the longer string of keywords. if you want flexibility, just use "blue widgets" and ensure that it is on 'Phrase Match' that way if some one searches and uses that term, but even additional words, your ad will show.
There are other ways to focus on London. Happy to discuss, or, i'm sure that there are threads to same effect.
"Using the three key words independantly will trigger fewer <em>quality</em> results" (oops)
I highly recommend organizing around small groups of keywords all with a sharp focus on one theme - and a well written and highly focused ad on the exact same theme.
That said, here is my take on your question:
The keywords below represent a focused theme - though I might add a few more (up to 20 total perhaps) so long as they are on-target:
- blue widgets in london
- london blue widgets
On the other hand the keywords below do NOT represent a theme at all. They represent three evidently unrelated things which are also extraordinarily general:
So, I would say use groups like the former, and not like the latter.
Thanks for the answers...
I was wonderning about the relevance of the keywords in both the cases, but I have done a couple of experiments, and yes... the right choice still the first ;)
|though I might add a few more (up to 20 total perhaps) so long as they are on-target |
Don't mean to hijack the thread, but AWA could you please elaborate on your advice about "up to 20"?
I read that the system uses up to 50 "top" (whatever that means) keywords in an ad group to define the theme.
So it's clear that adding more than 50 keywords is pointless.
But why the limit of 20?
If I can find 50 keywords that are tightly focused (just as focused as 20 would be), should I include up to 50?
Or is it better to use up to 20 keywords for some reason?
I can't speak for AWA, but personally, I think more than 20 is going to almost certainly dilute the focus of the theme. If there are 50 or more keywords/search phrases that describe it, it probably needs to be tightened further for optimum performance.
What I've found to be one of the most successful ways of organizing the content network is to play the game 'taboo' with your theme.
With taboo, you basically describe the theme without using the theme's actual name. If you look at an article, what other words would show up on that page in describing that product/service.
For instance, if you saw:
you know that the page probably contains information about the iPhone. You shouldn't necessarily use all the words you could come up with in taboo in describing that theme, but it will give you a good start.
Let's say that i would like to sell an iPhone in London an AdGroup may be:
Apple store in london
Buy mobile phone in london
new iphone 3g
Can you give me an example with 3 / 4 keywords ...?
just to better understand...
PS: I have also heard about the AdGroup title relevancy, is it true?
You definitively want to use keywords in content just as you do in search. That is, you don't want to have one-word keywords. You wouldn't need too many to establish a theme, 20 as AdWordsAdvisor suggest should be plenty.
There is one thing however I've always wondered about and maybe AdWordsAdvisor can answer this. I suggest making all your keywords phrase matches. Maybe it wouldn't make a difference but to me it seems logical to do so. I want to target sites with those exact phrases, not broad match them. Or does Google treat your keywords differently for content?
It's not looking to match sites by keyword (i.e. find sites who includes your exact search phrases on their pages)
It's using your keywords to figure out a theme, and then placing the ads on sites it deems to be within that theme - whether or not they include the exact keywords or search phrases you specify in your campaign.
If I have
blue widgets in london
as a keyword, my ads won't *only* show on sites that have that phrase on their page.
Therefore, match type is irrelevant.
|I think more than 20 is going to almost certainly dilute the focus of the theme |
I keep hearing that same argument over and over. But that's simply not true. There are plenty of cases where you can find more than 20 phrases that are still tightly focused. And splitting them up would pretty much duplicate the same exact theme.
I can't think of any, myself. But all you can do is try for yourself.
|I can't think of any, myself. But all you can do is try for yourself. |
Give me a phrase?
> as a keyword, my ads won't *only* show on sites that have that phrase on their page.
I understand that netmeg. And I wouldn't want to target specific keyphrases. But I do want to target themes and it just seems that two or three-word phrases do that just fine. I don't want to target each of the words in the keyword which could be anywhere on the page. I want to target pages with that common phrase, just like I do in search. Does it make sense? But if Google finds pages with what it considers the main theme of your keywords, I guess it's irrelevant.
> Give me a phrase?
I've got a client that sells a screen capture software. I've split his content campaign into three groups based on different variations of the main keyword: "screen capture software", "screen grabbing software" and "screenshot software".
Netmeg, do you think I'm splitting hairs here, would it make a difference?
I would probably put those three in a single ad group.
I'd recommend running (as does Google) 10-15 of your highest volume terms per category ad group. Highest volume terms will cover everything else.
For instance, different categories are widgets and placemats.
You'd want to create the following ad groups:
OK, let's try with your specific example "screen capture software".
Here is roughly 50 keywords:
demo screen capture software
advanced screen capture software
simple screen capture software
commercial screen capture software
good screen capture software
shareware screen capture software
quality screen capture software
easy screen capture software
cross-platform screen capture software
freeware screen capture software
popular screen capture software
cool screen capture software
screenshot screen capture software
windows screen capture software
best screen capture software
download screen capture software
professional screen capture software
easy-to-use screen capture software
fullshot screen capture software
custom software for screen capture
multimedia software for screen capture
popular software for screen capture
great software for screen capture
free software for screen capture
good software for screen capture
best software for screen capture
all-in-one software for screen capture
screen capture software programs
screen capture software reviews
screen capture software shareware
screen capture software download
screen capture software downloads
screen capture software utility
screen capture software freeware
screen capture software for recording
video screen capture software for recording
award-winning screen capture software
capture software for recording screen
software screen capture program
software for screen capture
powerful screen capture software
video screen capture software
free screen capture software
screen capture software
free screen capture software freeware
free screen capture software download
free screen capture software downloads
free screen capture software utility
screen capture software for professionals
screen capture software for pc
screen capture software for mac
screen capture software for windows
I didn't sort through them. So there might be some garbage there.
But please, can anyone tell me why I would want to limit the ad group to 20 keywords, instead of 50?
Is there any downside?
I can split it into, say, windows screen capture software vs. mac, etc. But then, I would still be left with a question of whether I should look for 50 keywords or only 10-20, for each of those split ad groups.
That's sure not what I would consider to be a tightly themed group of keywords:
PC/Windows and Mac are different themes
Free, shareware and professional are different themes
Programs and reviews are different themes
All those descriptive words like powerful, award-winning, cool, good, best, popular - that's just waste. None of that will help you on the content network.
I'd take that list down to maybe five or six *at the most*. And do well with it.
Have I got this right?
I create focused Adgroups using specific keywords
- Google reads my Adgroup and thinks "ah, he is advertising widgets"
Google then searches the content networks for sites about widgets
- displays my ads on the sites about widgets regardless of the words used on the pages
So the critical factors are:
- Google's understanding of the theme of my adgroup
- Google's understanding of the theme of a publisher's website
- the words used on any one page of the publisher's website are *irrelevant*
> the words used on any one page of the publisher's website are *irrelevant*
No. Keywords on the page is what Google will determine what that page is about. How else would they do it?
The keywords are not irrelevant. But they don't need to be an exact match on the page.
For example, if I have a keyword for football uniforms in my campaign, that ad may well show up on a page about football, where the word uniform isn't present.
There are no absolutes here - you have to test it and use what works for you. It takes time and effort. Mostly time.
OK - thanks LucidSW and netmeg
|It's using your keywords to figure out a theme, and then placing the ads on sites it deems to be within that theme - whether or not they include the exact keywords or search phrases you specify in your campaign. |
So maybe "irrelevant" was over the top. But my point is this - if the theme is the main connector than why don't we just use one word/ term and be done with it?
Why bother loading up adgroups with lots of terms if the category/ theme of the term (as determined by google) is the factor that Google uses to determine where to display?
netmeg I take your point about testing - but I get frustrated by the "just spend some money and you'll find out what is in the box" (not you, Google) approach. I would rather have some information before I part with the hard-earned $.
|Why bother loading up adgroups with lots of terms if the category/ theme of the term (as determined by google) is the factor that Google uses to determine where to display? |
I don't. And that's why Google themselves advises a few keywords per ad group, tightly themed/focused.