You can find a quick index to the entire guide here. Posts will be added as we write them.

So let’s begin with the process of creating a niche site itself.
Everything start with finding some “seed keywords”
Let me tell you right from the beginning that this first step has always been for us the hardest of the whole process of niche site creation.
Not because we haven’t got our share of positive results, but mainly due to the fact that I think that it’s more fair to say that we have found effective workarounds, that to say that we have found a “formal” solution to it.

What are Seed keywords ?

Seed keywords are keywords that you are going to be using to generate a lot of other related keywords.
This “expansion” process of going from one (or a few) keyword and ending with a lots of them, is something that you can easily try for yourself with the Google Keyword Tool.

But I don’t want to get too much into how to use the Google Keyword Tool, or alternatives to it. That’s something that we are going to talk about in the next post of this series.

The main purpose here is to understand that in order for you to be able to get your “expanded” list of keywords, you need to start with one keyword (at least) first.

What to look for in a seed keyword ?

There is no rule written in stone here, but here’s a few pointers worth to consider when choosing a seed keyword.

Commecial intent

This is probably the most important point of all. This is something that you will re-check when you make your final keywords selection, but with a slight twist to it.
At this point in the process, you don’t need the keyword to be overly commecially driven, but you need to be able to make the connection between the keyword and the intent to buy.

Let me give you a few examples:

Extreme commercial intent: “where to buy a Ford Fusion in Los Angeles” . People searching for this are already thinking in buying, they know what they want to buy, and they are looking for a dealer.
High commercial intent: “Ford Fusion features”. The person who search for this is still very specific, We don’t know if he/she is already thinking in buying, but depending on your site monetization strategy it could still be a valuable visitor. As an example, he/she could be really well predispose to click on ads.
Medium commercial intent: “buying a car”. Very vague. Could easily be willing to click on ads. Probably not good for an affiliate site.
Low commercial intent: “pictures of cars”. There’s no reason to assume that the visitor has any intention of doing anything else that looking at some pictures.

Anyway, the consideration of the commercial intent of keywords will be a much more important factor when we reach the stage of selecting the final keywords. When you are choosing your seeds, there’s still some room for error since low commercial intent keywords could expand into some high commercial intent keywords.

The keyword should be related in some way to something that people could search online for, while having some buying intentions behind.

Something to write about

Your seed keywords will generally determine what your content will be about.
Try to select keywords that allow you to write about. Creating content for some topics is far more easier than for others.
Try to get a rough idea of what will be the angle of the posts on the site.

Long tail keywords vs Niche markets

This is an often overlooked distinction. The whole idea of the “niche” concept is that you are trying to find a market to target where demand (traffic) might not be that high, but the offer (competition) is low.

A keyword that meet that traffic and competition conditions, may be a frequently searched term by a small sized group of people (niche market), or it can be a infrequently searched term, by a large sized group of people (long tail keyword).

With that in mind, when possible, go for the “niche market” alternative.

Let say for instance that you find 2 keywords with the exact same traffic and the exact same competition.
One is: “old 45 rpm record players” and the other one is “multi gear bicycles”.
The first keyword, clearly targets a small group of people, who are music enthusiast that also loves the old sound of vintage record players.
The second keyword instead, seems to be targeting a much larger group. It just happen that the majority of people search for “multi gear bikes”, changing the word “bicycles” for “bikes”.
Google is extremely good at making the right keyword replacement when is more or less obvious what you were looking for.

Where to look for ?

I don’t have a definitive answer. But it seems to me that a large percentage of Internet Marketers use Amazon or Ebay (or some other retailer) categories.
That is ok, specially if you are going to run an affiliate site, but keep in mind that many people will be doing the same thing that you did.
Google Trends: Same thing, good place to find ideas, but plenty of people looking there.
Offline world: You can get many ideas from looking at the shops in your area. Keep your eyes and ears open, and a lot of potential seed keywords will come to your mind.
Go local: local Internet searches are taking over the IM arena. A highly competitive keyword could be extremely within reason when you narrow down the geographic targeting of your content.

Our approach

I admit that we are guilty as charged. We have found a workaround for this step. Is pretty simple: We just let our software go viral. Let me explain.
Suppose that I think of the keyword “sport cars”. That keyword will get “expanded” by our software into the roughly 800 keywords that Google returns to you in the Google Keyword Tool.

Our software is set in a way that when it has nothing to do (let’s say we are not running any process) it takes each of those 800 returned keywords, and uses it as a seed for a new run.
So potentially, in this second run, you get 640.000 keywords.
Of course that you get a lot of repetitions, but out tool discard the repeated results, and keeps the new one. Even with lots of repetitions, it’s extremely likely that we get over 100.000 keywords from a second run.

It is true that sometimes most of the results will be equal to the ones received in the original execution, but who cares, it takes almost no extra time, and even if you get 2 new results (you are likely to get over 20) you are expanding your database at a crazy rhythm.

After that, we check for the top 10 rated keywords keywords and we always find something there that is interesting enough to create a niche site around.

Please, tell me what you think, and what strategy do you use to come up with new seed keywords for your niche sites.