Keywords: research is the key

When starting an online campaign, the most important thing is to do some keyword research first. This will include looking at the business value of the keyword you are considering. Two good examples of this are the tattoo market and the dream analysis market, both of which have quite a few searches, dream analysis even more so. There are quite a few searches there, but the business value of that particular keyword is low. Most people are looking for a free dream analysis, and they are not actually going to buy anything.

You can do your research to flag traffic and flag low competition and say, okay, here’s a keyword I can go after. Then when you hit the market and try to monetize, you’re really going to have nothing.

I think a lot of people are scared away from markets that have competition, and if anything, I think Jeff Johnson talks about this, you want to go after competitive markets because it shows there’s money there. All you have to do is a little better.

We do a lot of testing on Market Samurai with the traffic component, relatively recently Google started providing some kind of information about how much a particular keyword is getting. Through some of the testing I’ve done, I see massive discrepancies. For example, you can have a number one position ranking for a keyword and not even get a tenth or even much less than that click on that particular keyword.

I think ultimately that’s the thing with testing, you take it with a grain of salt and look at multiple sources. As with stock trading, you don’t just buy something based on one signal, you look for multiple signals to say the right thing before taking action.

Most people are so tied up in analytics paralysis, and particularly with the product that we provide, we provide data that people can become so obsessed with and it’s good that it provides some discipline and some rigor to what can be a Very risky business if you don’t know what you’re doing.

I like to do an AdWords campaign first and confirm the data. I know that AdWords data should be a percentage of the total number of searches, and AdWords will also tell you the number of impressions your ad got. So if you’re looking for confirmation, you could spend a little money to make sure the traffic is behind a particular keyword. Proper keyword research is key.

I certainly do AdWords and pitch it to a site and test for business value and conversion and traffic before investing heavily in that market.

So one of the first stages, both pay-per-click and SEO, is obviously to do some keyword research. With SEO, obviously, you can take those keywords and load them into your pay-per-click and start monitoring conversion and that sort of thing and tweaking things.

Once I’ve done the research, I consider how many keywords to choose and where to start.

I try to choose a small subset of keywords that are tightly clustered semantically with each other and that meet certain commercial and traffic competition filters that I impose on them. So basically I’m saying I’m looking for keywords that have a decent amount of traffic, so they’ll be worth my time, with a decent amount of business value.

Obviously it’s going to be how my competition threshold is based on what my SEO asset base is in terms of my other sites and my ability to get links. So I like to think that this group of keywords essentially becomes categories.

If it’s a blog on my site, I usually have six to twelve important keywords or something behind my site. But there is definitely a semantic clustering of those keywords. They are related to each other. That’s really one of the cool things behind the Google Keyword Tool. In fact, they are giving you semantically relevant keywords and that is very important.