Forum Moderators: buckworks & skibum

Message Too Old, No Replies

Keyword variations in Adwords

         

Flower_Girl

7:42 am on Jan 12, 2007 (gmt 0)

10+ Year Member



Hi. I am new to Adwords and I seen something which really confused me. It sort of made me realise why some people are so succesful and why I seem to be failing in mine.

I was checking out positions for an adword campaign. I put in my keyword as "blue widget store", so hopefully anybody who searches for any terms like buy blue widget store will get my ad. This is right, right?

Anyway, i was checking out the ads and positioning and I realise something. There are these few sites, and no matter how you type your keyword, they ALWAYS show up. So you can put these words in google, like

bluee widget store
ablue widget store
blue widgeti storee
bluef fwidget storeg

any variatins like this, and their ads ALWAYS show! How do they do it? I am sure they did not think up of all these combination and put them all as their keywords? Or did they?

Kings on steeds

9:31 am on Jan 12, 2007 (gmt 0)

10+ Year Member



Flower_Girl, I think I understand you, but it was a long night. Google has three main matching options,

Broad - any variation of this keyword, Widget, Blue Widget, Red Widget ect..

"Phrase" - any keywords before or after your keyword, but you keyword must appear in the order it is listed.

[Exact] - Nothing before, Nothing After just that word!

For keywords on broad match there is something called extended broad match, this shows you for related terms & misspells. if you think about the traffic google receives, thats a lot of misspell data! so they do well!

Hope this helps!

Alan

[edited by: Kings_on_steeds at 9:38 am (utc) on Jan. 12, 2007]

Israel

6:18 am on Jan 16, 2007 (gmt 0)

10+ Year Member



Hi Flower Girl,

Since KingsOnSteeds is tired, I'll add a little bit about the topic:

How do they do it? I am sure they did not think up of all these combination and put them all as their keywords? Or did they?

I'm old to Adwords but have been a bit out of the loop for a while. Things have changed, no doubt, but I suspect this part remains true.

At the outset of a new campaign, even your broad matches will mostly only show if the exact version is typed - strange as that seems.

Over time, Adwords will start to 'test' you out with exapanded versions of your broad matches to see how they fare. If the variations generate clicks you'll gradually become eligible to show for those variants and to 'expand' more and more.

I see this happen in my perennial campaigns which have been around for over 3 years as opposed to a campaign I started yesterday.

So unfortunately, longevity of a campaign/ad/keyword counts for a lot in Adwords (as well as the SERPs).

Fortunately, you'll gain that longevity over time.

Expanded broad match can be both a blessing and a curse. Computers don't understand a lot of things within their context and an old campaign or ad's keywords may show for a lot of synomyms and phrases totally unrelated to your product or service. You eventually (perhaps even now) may need to include a lot of -keywords. Right off the top: -free if you're not giving the blue widgets away, -used if you don't sell used widgets and on and on.

Adding a whole bunch of keyword variations that are unlikely to generate impressions more than once in a blue moon is normally a useless endeavour. Adding clever variations by asking friends how they'd search for your product or by just plain sitting there and pondering can yield good results.

So no, those people did not add all those mis-spellings and variations to their keyword lists. Chances are, through time, they've managed to generate a sufficient CTR with blue widgets or maybe just widgets that they've qualified to show for almost any variation.

BTW, only those with deep pockets can afford to absorb all the irrelevant clicks the'll garner simply by advertising on 'widgets', especially when they don't even have blue ones.

Also, I've found it near impossible since the introduction of the quality score to include mis-spellings in my keyword lists without paying $10.00 for them. Don't know how others have fared.

For now, I'd concentrate on the most specific terms that describe your product. Include plurals, articles, prepositions, etc. in your keyword lists plus any obvious synonyms people are likely to use. Use the Thesaurus for ideas, think about the fact that perhaps younger people use a different name for your product than older people. Download a completely freeware product called Good Keywords which simply shells the YSM/Overture search data (which is public) to get some ideas of how people search. They have even have a newer version which does better than that. Keep in mind, the YSM/Overture search data, when shown, omits articles, pronouns, prepositions, etc. IMHO, you'll want to include most of those at the beginning.

Then, in time, you'll find you may have to winnow down your keyword lists and use -negative keywords liberally to keep out irrelevant "expanded broad matching" that might only bring down your CTR.

HTH,

Israel

Flower_Girl

2:38 pm on Jan 16, 2007 (gmt 0)

10+ Year Member



Thanks guys, your explanation is spot on :)

RonnieG

5:45 pm on Jan 16, 2007 (gmt 0)

10+ Year Member



Also, I've found it near impossible since the introduction of the quality score to include mis-spellings in my keyword lists without paying $10.00 for them. Don't know how others have fared.

There are two or three misspellings and variations in my vertical that are very common, to the point that G may in fact count them as normal spellings, so my bids for those variations are still slightly lower than for the proper spellings.

In some verticals, G also uses a list of synonyms for some keywords in expanded broad matches, not just misspellings. I have been quite surprised at some of the synonyms that can trigger my ads, and have added -negative keywords for a number of those. Since I have highly targeted ad groups for many of the synonyms as well, I also find that I have to use -negatives in nearly every ad group to keep the QS for each ad high by excluding the synonyms for which I have other unique ad groups.