The keyword research phase is useful for knowing if your website's niche is worth diving into. Pat Flynn recommends that you NEVER go into a certain niche if you're not getting at least 3000 local searches for your main long-tail keyword. Do you know what your niche is about? When you go to Google Adwords, are there enough searches to make this site worthwhile?
Once that's done, you can honestly settle down on doing intense keyword research. So long as you write for your audience and you KNOW that the content that you are writing is helpful for them, you should have good long-term success. For example, if your site is about losing weight for the elderly, just focus your content around helping that specific audience. Maybe what exercise equipment elderly people could use, what foods they should eat, maybe create a community of elderly people who want to lose weight, etc.
Don't worry about looking for specific keywords that match your article. Based on my experience, you'll start getting a ton of searches from random long-tail keywords that have zero presence on any keyword tool.
You should also determine if you want a static site or a blog before you go any further. The format you choose can make a big difference in your performance.