The Three Different Approaches to Keyword Research

Is there a single way to research keywords? Or are there different approaches to keyword research?

Once you’ve selected your niche, and understand your audience, it’s time to select the keywords you will target in your campaigns.

And based on most of the training, there seems to be a single approach, which is not the case.

I call it the “Analytical” approach to keyword research. This is the most commonly taught method: you select your favorite keyword research tool, like Market Samurai, Traffic Travis or my favorite, Ultimate Niche Finder, enter a seed keyword, expand it, and start looking for related keywords that met such and such criteria (like number of competing pages, search volume, how optimized and strong are the top X results, etc).

And when you find enough of them, you build a site targeting these keywords.

Don’t get me wrong, I like this one and it’s the one I follow in most of the cases, as it has its advantages:

You have a clear idea of what keywords you are targeting and thus can optimize your site or page(s) for them, what the people searching those keywords may be really looking for, and a general idea of how strong your competition is, etc.

But it also has some drawbacks:

  • Most of the keyword research tools rely on Google (and their AdWords keyword tool) data to give you an estimated search volume. And Google sometimes gives inaccurate information about that volume.[break][break]Whether it’s because they are calculating averages throughout a long time, and they are not always updated in real time, or it’s because they simply skew the results towards keywords that give them more profits on the AdWords platform, no one knows for sure.[break][break]This can lead to some undesired surprises, after you rank for a keyword and don’t get as much traffic as you expected.
  • Even though you research your competition, and you can get a fairly decent idea of how competitive a keyword is, you can never be 100% sure about your ability to rank for a keyword.[break][break]So, you can invest a lot of time and effort in a supposedly “easy” keyword only to find later that you can’t make it to the top 3 and get sufficient traffic from Search Engines.
  • And finally, even if all goes well and you are able to rank for a keyword, and you do get the expected traffic, you don’t know in advance if that keyword will convert into sales.[break][break]The better you have researched your niche, the more likely it will be that you craft your messages appropriately and turn that traffic into sales, but this is not an exact science in any case.

So, are there any alternative approaches to keyword research?

There are in fact two:

The first one is what I call the empirical approach to keyword research. The idea is to create a sufficiently large site with enough posts about a certain topic or niche, without worrying too much about keyword research, seeing what long-tail keywords you rank for naturally, and then strengthening your position on those.

  1. The first step can be done either by writing the content yourself, outsourcing it, re-writing PLR, by using auto-blogging plugins like Digi Traffic Multiplier or WPRobot (with modules to grab content from almost every possible source), or by accepting guest posting.
  2. Then, promote it and ping the pages, possibly using some automatic syndication tool like WPSyndicator or Digi Auto Links.
  3. And in future weeks see which keywords and in which posts or pages you start ranking for (I use the Rank Tracker Plugin for that, very handy tool!).
  4. Finally, strengthen these pages’ on-site and off-site SEO factors for these keywords, or even create a completely new site targeting these (if you’ve proven that you can rank for a keyword in a medium quality site, it will be much easier to rank for it in a better quality, ultra-targeted site).

This way, you are removing guesswork out of the equation, as you are focusing your efforts in keywords you know you can rank for, and that send you a certain amount of visitors.

The drawbacks are that, first, you have to create the content first, and second, you won’t typically rank for highly competitive keywords using this method, as you are not optimizing the site or focusing your efforts in them. It’s an statistical approach: out of potentially hundreds or thousands of posts, you’ll likely rank for a bunch of long-tail keywords (which you can see in the Rank Tracker Plugin’s dashboard, or even have them sent to you via e-mail by the plugin) you would have never even though of.

And of course, finally, there’s a middle-ground approach:

  1. You do some analytical research, and select a decent (but not overwhelming) number of keywords you’d like to rank for.
  2. You promote the corresponding pages or posts, and try to get them to the top 3 positions for your target keywords.
  3. But, at the same time, you monitor additional keywords you are (unwillingly, and possibly unknowingly) ranking for using the Rank Tracker Plugin.
  4. And once you start seeing keywords you rank for, you either create new posts specific for those keywords, or you “refocus” your existing post or page to rank higher for those keywords they are already starting to rank for.

I’ve tried all three, and I personally like the last approach the best, as it is the one that best lets you adapt to the reality of the search engines and how they see your site.

What do you think of these approaches?

BTW, if you’d like to know more about different approaches to keyword research, I strongly advise you to read Keyword Reality and Keyword Needle, by Chris Munch, or the Keyword Shotgun Method by Gavin and Jake (from the “SEO Training Alliance”) where they all talk much more in depth about these topics and gives you specific methods you can follow to do your “out of the box” keyword research.

Let us know what you think below!