I understand the marketing basics on keywords and the use of tools like Google Insight, Keyword traffic estimator, and others.
My concern comes from a product that people usually don't search for. If people don't search for this product directly, how do you figure out the keywords. Is there any science to this, or just trial and error?
Update: Answer to a comment requesting an example.
I can't give the product directly, but I will give examples of similar situations.
One example that comes to mind is "Web Analytics". People that don't have this, will probably not look for those keywords, they will probably not even know they need analytics.
I hope this helps.
I know that there is some science or art to this. But this is what I do.
Then I go to Google to test them out.
I keep this all on a spreadsheet with little hash-marks. And the ones that seem to emerge the most-- well I circle those (with a pencil, on paper). I put them together on a list and show them to the client. Then we start the process of writing/editing copy, article descriptions, meta-tags for each component of our site. We feed them to the person who will be doing the ad buys over on Google and where-ever. And then, because we don't believe in them too strongly we respond to what the market tells us as we move forward. Tweaking, changing and learning.
I am sure that the people who charge a lot of money for it do something completely different. But I always seem to get pretty close to the same results. :)
If you're talking about PPC keywords then it's a much more precise science than the black arts of SEO. I say this as a person who has a lot of experience with both, but I generally prefer PPC when I can use it.
I would compile a list of terms that I think describe the product, then go search for those keywords to see what you get. If you can't find something similar to what you're selling, then try different keywords.
When you start getting matches that are close to what you're selling, then make note of those and start creating your PPC campaigns based around the keywords. Then refine them until you get a group of keywords that convert well.
It would have to be a VERY rare product to get no searches on it at all. It's partially trial and error, but you do learn quickly from your mistakes and you should be able to find the trail that leads people to your product.
I'm an SEO (no good at any black arts, unfortunately) and it's normal for SEOs to use PPC campaigns to gather keyword data. Actually, as an SEO consultant, I typically find that my clients have tried PPC in the past, and I always try to get my hands on their keyword data to use it as a starting point in my SEO keyword research.
Another thing to consider is what keywords your competitors are targeting. If you think your service is so new and original that you don't have competitors, you might be right in the business sense, but in the search engine sense, someone is always going to be competing against you for ranking in organic results.
Also, I'd recommend creating content for your site that's useful, interesting and relevant to the demographic that you want to become your customers. If you can drive traffic on topics of similar interest, you can market in that space. Start a blog.
Just answer your web analytics question.
Yes, lot of people do not know web analytics and it is impossible for them to search such keywords.Web Analytics is still a good keywords by the way, since it is getting great progress recently.
Now back to the answer and do an out of the box thinking. Web analytics involves traffic, visitors, keywords, on line marketing. why do not you try to optimize those keywords? Not necessarily those big keywords though. Long tail keywords like monitoring traffic will do the trick.
Did answer your question? I think I did. Good luck