SEO 2020 FORUM

[alex]

Hi ! Alexandra here,

Before I ask my question, I would like to thank Charles for giving us a comprehensive free search tool at:
http://www.seo2020.com/seo-tools/keyword-research/keyword-research-tool/

This posting is to ask Charles, Russell, and others who have real extensive experience in these matters, about some nagging keyword questions which are delaying my business.

Now getting lots of data has not been a problem for me [except for the wide disparity of the results], my problem really is in the interpretation of this data. Or to be more specific, my problem is what data constitutes great keywords that will get traffic and monetize well.

Based on the nuggets I have accumulated along the way (in my 9 week career ) from surfing different sites and blogs and forums, I have come to the following conclusions:

1. You want a main keyword that has at least 1,000,000 in gross traffic per day – this is not the traffic you get but rather total traffic generated by the keyword for all websites.

2. If you are selling affiliate programs, then you want low cost keywords to advertise on (if using Adwords) and these tend to be strings of 3+ words.

3. If you want to generate Adsense revenue, you want keywords that have a large number of advertisers and base your opinion on the average cost per click – not the highest bid as it will not be published by you. If you want to arbitrage with Adwords, you would bid on lower cost synonyms, longer word strings, and misspellings.

Your opinion on the above would be greatly appreciated.

Now as to the interpretation of results from different keyword generators and stats, I find a lot of difficulty in interpreting the results.

Problem 1. Each database source of data for different keyword tools has vastly different results for the same keywords. How do you reconcile that?

Problem 2. Wordtracker uses KEI – Sumantra Roy’s baby - as a measure where a KEI of 100+ is ideal to build on. My thinking is that if the competition is as low as 36,000, that is still formidable amount of traffic to overcome. I see a shortcoming there in using KEI, tell me why I am wrong. Now you may say that silos will overcome that in time with natural traffic – I say in that case that silos will eventually overcome anything but that doesn’t help me today. I am not badmouthing silos here as they are the religion which bring us together. But to put things in my favor, I should consider other ways to judge a keyword. Any thoughts? Also, if you can find them, some keywords that have less than 500 competition are around but then demand of 50 per month matches that. Once again the answer may be in longer strings. Please elaborate if you know about this from experience.

I would really appreciate some answers from people who have extensive experience in these matters.

Thank you in advance,
Alexandra

[jeremy]

Hi Alexandra,

So far what I have done is to choose a niche
where I can review physical products.

Then I go here:

http://inventory.overture.com/d/searchinventory/suggestion/

I like to see a niche that has lots of keywords with categories,
brand names, and model numbers.

Then I go here:

http://uv.bidtool.overture.com/d/USm/search/tools/bidtool/

to check the ppc, I like to see a top ppc of $1.00 or more
but at least over 50 cents is fine to.

I do all my keyword research in wordtracker and I get
competition results for google only. Why, it is the most searched.

Then I choose 5 to 15 main menu categories with high demand,
those pages are themed with less competitive keywords 100
or less.

I start writting articles with SPECIFIC keywords with competition of
10 or less and work my way up. I do look at kei, but I also
look at competition, if the kei is 200 but competition is 5000
I'll use that keyword later.

The only diffeent now is I will structure my sitemap according
to the Plan and I will link together related articles in a silo
structure.

Pretty simple.

Jeremy

P.S. I think it is a good idea to choose one keyword research
tool and stick with it other wise your comparing apples to oranges.
I've been using wordtracker successfully or 2 years.
Your own sites web logs should also be used.