I have been practicing keyword research across multiple sites for the past eight months. I have done it for niche sites, for this blog, and for my clients. Most recently, I have been doing a lot of keyword research on niche sites for my mass niche site project.
If there is one thing I have learnt in the last eight months, it is that most people who are trying to build profitable niche sites are going about keyword research in entirely the wrong fashion. As soon as you start overcomplicating the process, your effectiveness in choosing profitable keywords disintegrates.
Here’s a simple fact – finding low competition keywords, when systematized properly, is in fact a relatively straightforward process. If you want proof of that, look no further than the likes of Spencer Haws or Justin and Joe – guys who are producing niche sites on a grand scale. If keyword research is that difficult, how are they able to keep rolling out tens (or even hundreds) of sites every single month?
Picking profitable keywords is a process that I want to cover in great detail in the future. However, I am only going to do that when I have ample evidence to back up my own methods. In the meantime, I want to draw your attention to three reasons why your niche site keyword research strategy if ineffective. If you take the below points on board, your likelihood of picking profitable keywords will be greatly improved.
1. You Are Trying to Rank 1st
There are ten available spots on the first page of Google, and each spot will attract a certain percentage of clicks, depending upon numerous variables. I wrote an in-depth post on this topic over at Think Traffic, which I recommend you check out.
Let’s assume for a moment that 40% of searchers click on the 1st result, and 20% click on the 5th. You’re looking to find a keyword that will send you 1,000 exact match visitors per month. So a keyword with 2,500 exact match results for which you can rank 1st in Google will do the trick. But a keyword with 5,000 exact match results for which you can rank 5th in Google will also work.
In fact, you could argue that the second keyword is a better one to target. With the first, there is a definite theoretical ceiling of 1,000 visitors. But with the second, if you exceeded your expectations and ranked higher, you could attract up to 2,000 visitors.
Stop focusing on the 1st spot of Google – there are 10 up for grabs on the 1st page, and they can all send your website traffic.
Whilst I used to use Market Samurai for keyword research (old reliable), I now use SECockpit (hugely powerful, but buggy – I will be reviewing this tool soon). I actually use SECockpit’s method of dividing the first page of Google up onto three “sections”:
1. 1st-3rd spots
2. 4th-6th spots
3. 8th-10th spots
SECockpit calculates that each section will attract a fixed percentage of traffic. It isn’t perfect, but it is impossible to calculate the precise level of traffic you will receive from any given keyword. For my purposes, SECockpit’s estimates are good enough.
Once you have split the 1st page of Google into three parts (and taken the differences in traffic into account), you can analyze the competition in each section independently of the rest of the page. This allows you to consider keywords for which you do not expect to rank 1st in Google.
2. You Research Only One Keyword
I was guilty of this until only recently, and it is a really crazy thing to be doing. In looking to build a niche site, I would go out in search of a keyword. Once I found a keyword that I thought would do the trick, I would go about building the site.
This is wrong, wrong, wrong! No two keywords are created equal, and you want to have a considerable backlog of researched and ready keywords before you decide to build a site.
Justin and Joe of AdSense Flippers say that for every one keyword you build a site for, you should have researched 50 other potentially viable keywords. This may sound over the top to you, but the fact is, the more keywords you research, the better chance you will have of finding more profitable ones. If you only research one keyword and then build a site, you are reducing your chances of it being a success.
There is always strength in numbers, and keyword research is not an exception to that rule.
3. You Don’t Score Your Keywords
This follows directly on from my previous point. You must have a way of scoring your keywords. If you are going to research 50 keywords for every one that you decide to base a site upon, you need to know which of those 50 is the best one to target.
By no means is this an absolute science, and your scoring system is bound to evolve as you gain experience, but something is better than nothing. I personally score my sites based upon a number of variables, focusing on offsite and onsite SEO. I will probably reveal my scoring system at some point in the future, when it has been adequately proven to be effective.
In the meantime, you need something. I wouldn’t worry about getting too anal with your system to start with – just find a way of ordering your keywords, see how well your system reflects the performance of those keywords, and adjust accordingly.
Do YOU Have Any Suggestions?
I have covered above what I consider to be three vital areas of keyword research that are neglected by many niche site builders.
But there is certainly more to be said. For instance, there are various common keyword research mistakes that you must avoid.
So if you build niche sites, please feel free to reveal your tips in the comments section below! And if you think this article would be of help to your followers, please share it using the buttons below.
Creative Commons image courtesy of Brooks Elliott
Andre Garde says
Hi Tom,
A good work to score your keywords is to use a simple formula to calculate the anticipated revenue each month if you hit the top spot. That’s essentially what I’ve been doing and it works out rather well. Most people will target $1/day, which is reasonable.
With the keyword research, another nice trick is that you can target multiple keywords per page or post, as long as you do your onpage SEO properly. I experimented with this for a couple of my sites and it works out well, since you can own 1, 2, or 3 spots with just one piece of content.
Andre
Tom Ewer says
Hi Andre,
Perhaps I wasn’t specific enough – by scoring, I meant by competition, as well as by income. I think both factors are important. But you are absolutely right in recommending that you produce an estimate of earnings. Spencer Haws did a great post on that topic.
Interesting idea RE targeting multiple keywords…I like your thinking. Great suggestion – thanks! π
Tom
Jason says
Keeping score is a definite must. I signed up for SE Cockpit for one month. Toward the end of the month, I exported all my keyword searches/results. I imported into Access (or you could just cut and paste in one excel worksheet), slapped my secret sauce formula on the data, and presto! viable keywords (except for the huge step of competition analysis). I don’t really have a secret formula, or else I’d be selling it for some dollar amount ending in 7.
Once I get more sites under my belt and some good data, I’m going to try to develop some fine-tuned formula, but until the standard formulas that everyone throws around works (or just use the Adsense Value from SECockpit). This will at least provide a measure to sort and start working your way down the list.
Tom Ewer says
I always like to read your comments Jason because it’s like I’m reading through my own thought-processes π
We’re two very analytical guys by nature, and I think that is a very handy attribute to have when working with niche sites. Once you have your systems in place, it can quickly become a numbers-based game to a large degree.
Gregory Ciotti says
Hey Tom, one tool I’ve been checking out (but haven’t used recently, no idea what the team’s been up to) is SerpIQ, I like the interface post keyword search the suggestions can be adjusted and change on-screen.
Only free for a few runs though, I’ll have to see if it’s worth utilizing over Market Samurai, which is great but so damn slow sometimes.
Tom Ewer says
Hey Greg,
Looks interesting…I’ll have to take a look. Thanks for the tip!
Tom
Steve Wyman @ Propelbakclinks says
Hi Tom
Firstly I dont like SECokcpit as its way to expensive unless you do as otehr do which is subscribe for one month and extract as much data as your bandwidth will allow. And even then i think its expensive
Some people swear by it but they are typicaly linking to an affiliate earning which ofcouse is very attractive.
Ive spent years dong KW reserach both for niche sites and for offline business. Using google tools and pretty much every other tool ivetried.
The hands down winner to find keywords to build niche sites around is LongTialPro. Produces by Spencer @ NichePursuits.com its is buggy and has some issues in the work flow (like once youve done a run its hard to edit thats run) but i run it on my VPs server and its wild.
You can put in upto 5 keywords and get back 5 x 800 results. Set the filter and boom there are the 5-10 (typically for my extreme criteria) keyords which also have an EMD. (you dont have to filter youcan manually search if you wish.
Bit of sales pitch but ive found more amazing keywords in 2 weeks than ive fond inyears using other tools.
Dveloped by one of the best niche developers it delivers for me.
regards
Tom Ewer says
Hi Steve,
I can understand the argument against SECockpit RE cost. It certainly is pricey. My justification would be that the time saved in using the tool far outweighs the additional cost.
I have tried using LongTailPro on 3 separate occasions, but have never been able to get along with it. I really want to love it, as it is Spencer’s tool and he knows what he’s doing. My main issues are:
SECockpit is just so much faster. I can research say 10 keywords, export all of the data into an Excel spreadsheet, filter it by AdWords competition, CPC, and traffic, then by SERP competition, and I’ll have a list of viable keywords, all ready for final competition analysis.
Cheers,
Tom
Darren says
Hi Tom,
I think I fall into that category of overcomplicating keyword research. I would love to see what you do in more detail. I only recently purchased Long Tail Pro, and I still think I got it wrong…. (not the softwares fault, but the user)
Cheers
Tom Ewer says
Hi Darren,
It’s a process – the more you do it, the better you get. I am keen to reveal my technique, but only when I know it is at least a little effective π
Cheers,
Tom
Steve says
Hi Tom,
I find market samurai a bit frustrating. Yes it is feature rich, the competition data being excellent as you know.
But it is not very well organised for developing a target list of keywords at all. The tab-per-keyword seems like a bit of an afterthought and It is really built around 1 keyword at a time. A tool that combines the MS power around the ability to develop a list of keywords.
Not sure that the SE Cockpit buys have broken the mould there, I can see tabs-per-keyword appearing.
Great post as always though!
Steve
Tom Ewer says
Hi Steve,
I think you have to have tabbed keywords – I don’t see how else you could do it. Each keyword should be analyzed independently – there’s just too much competition analysis data to include several keywords on one screen.
SECockpit might go some of the way towards alleviating your frustrations by giving you a “Competition” bar, which is a rough indicator of the competition levels per keyword.
Cheers,
Tom
Steve says
Hi Tom,
Well you might be right there, if there is a better way we’ll know it when we see it.
In the meantime, I’d better take a closer look at what they are doing differently in SECockpit before making a call there. Its a bit of a plunge with no trail though, so got to be ready to invest some serious time.
Will be interesting to hear more about your usage and tips in future posts π
Cheers
Steve
Vince says
There are some excellent keyword research tools like Semrush, Googleβs keyword planner,Ubersuggest and Long Tail Pro . I am using Google Keyword Planner and Long Tail Pro for my keyword research work. I actually choose a lot of long tail keywords and some short tails with high searches and so far, it’s working fine for me.