Forum Moderators: open
Basic keyword popularity only reveals quantity and fails to make the distinction between general public and your target audience. There are programs that estimate good keywords based on popularity and competition, but again fail to make any direct correlation to your specific audience.
Site metrics reveal what keywords where used to find you site, but whoever found your Web site may not be in your target audience, or an accurate reflection of your entire audience. In other words you may have potential audience using keywords that you are not ranked for.
With a traditional marketing background, my instincts are to conduct market research. You know, sit down a bunch of people in front of Google and ask them to find us. But its hard enough to convince clients to spend money for the limited keyword analysis we currently do.
If you forget all the tools that may help for the moment, define yourself in generic terms.
Start off with single words preferably as nouns and then double nouns then pair these words (and short phrases) up with adjectives and in different arrangements.
It is likely that people will come to the same conclusion as you - at least a percentage and to re-confirm plug them into google and see if competitors appear.
The single/double nouns are likely the ones you would want to target but these can have enormous competition, so in the short term 3 word phrases are better or longer (to start getting a feel). You would be surprise at how wierd some search queries can be, but the search engine looks for the best result so in many cases you don't need to target.
Some people add geographical locations to their queries so having your physical address on each page can attract these searchers.
Currently I have 16,000 limited use keywords on the first page of results, although only use once or twice a day this is significant traffic in total and most were never targeted keywords.
In addition, these limited use terms have helped identify more common use terms without aid of any tools -- when people don't find what they are looking for they tend to re-define their search by being more specific.
Search Engines with spy technology can help you determine searching trends. Watching these are awhile can give a good appreciation of general or common search habits.
There are many out there but the only one I can think of off the top is Genie Knows [genieknows.com]. The search engine itself sucks (totally commercial) so their results are bias to who has the big bucks but searching habits can be easily seen in a short period of time. Genie Spy is in the middle of the page.
[edited by: fathom at 1:24 am (utc) on Sep. 12, 2002]
Another good resource is to hunt for "those counters" no matter what the product or topic and look at how people use searches at Ask Jeeves. It may only work for certain markets, and my impression is that it's heavily oriented to what women are likely to search for, but a lot of those are consumers, and you can get a good idea of the self-speak process that takes place when people search. It's not only picking keywords, but how the keywords are used in formulating phrases.
I find the odd time I actually see and talk to a client, I am making notes of his voiced "keywords" ;) and forget the clients real world comments and questions.
Currently I have 16,000 limited use keywords on the first page of results
This is a great strategy. How would you handle 16,000 keywords in practice. Are these content pages each written to be optimized for a keyword (or phrases).
Would you then submit each page to Google, and how about incoming links. It would not be practical to go for incoming links for all the resulting pages.
Appreciate your help.
ezGuy
