SEO Keyword Research

I don’t think it’s at all possible for me to over-state the importance of keyword research. Take a look at the following…

How much will my Adsense ads pay out per click?
How hard will it be to get ranked?
How many links will I need to build?
Am I better off buying an established site, or can I start with a new domain?
How many pages of content should I add to the site?

All of these are questions that get asked when you’re publishing a new site, and the answer to every single one of them is: It depends.

And what does it depend on? Ultimately, your target keywords. Let’s look at these again, with more clarification…

How much will my Adsense ads pay out per click?
Different terms are worth more to advertisers, so they pay out at different levels. Google gives you a good estimate.

How hard will it be to get ranked?
This depends entirely on the competition. Who the competition is varies based on what keyword you’re looking at. Keywords determine competition.

How many links will I need to build?
More than your competition. The competition depends on the term.

Am I better off buying an established site, or can I start with a new domain?
Depends on competition. The competition depends on the term. If all competitors are on sites that are 5+ years old, you probably want to buy one.

How many pages of content should I add to the site?
Depends on competition. The competition depends on the term. If all competitors have 100+ pages of good content, you better get writing. (Just kidding. Scrape/spin or buy the content.)

In short, your keyword research is the foundation process that will completely dictate how long and hard the road ahead is, as well as the size of the reward at the end.

Target “red dog sweaters for German Shepherds” and it should be a piece of cake to rank – but you won’t make much, if anything.

Target “credit cards” and you’ve got a long, hard road ahead. However, if you get to the top three – you can unload the site and retire.

Finding Your Basement

Personally, my main goal is finding keywords with decent volume, relatively weak competition, and good CPCs. I realize I just did some vague marketing speak on you there, so I’ll clarify these with hard numbers.

If I’m building a small MFA site, say 10 pages, I’d look for keywords that get at least 1,000 exact match searches per month with an average CPC if $2 according to the Google Traffic Estimater. To answer the question why, we need to do some math.

My data shows the #1 result in Google is now good for a reliable 25% CTR. So 1,000 searches a month means I can count on 250 visits from that terms when I get to #1.

Google Adsense pays out 68% to the publisher. 68% of $2 is $1.36. I want to end up at a dollar per click, and not all ads will be at the level of the CPC estimate, so I like to give myself a cushion.

I know based on my templates that I’ve tested that I can count on a CTR of 15% on my Adsense ads.

250 visitors a month with a 15% CTR is 37 ad clicks. 37 ad clicks at $1 per click is $37 a month. 10 pages gives me 10 primary keywords. So, getting ranked #1 for my ten primary keywords that fit the above criteria should yield, at a minimum, $370 a month.

That’s more or less my basement – the $350 – $400 range per small MFA site.

Now, keep in mind I’m not aiming all that high – 1,000 searches a month is *not* a high volume term by any stretch. However, with a $2 CPC, they tend to be valued keywords with some fairly active competition. You could certainly look at 1,000 searches a month with a 20 cent average CPC and likely have a much easier time ranking. It all comes down to how big a task you want to give yourself.

Picking a Niche

Seriously, the niche can be anything. People recommend that you pick something that you’re passionate about… but, meh. I make sites about stuff I have absolutely zero interest in and they work just fine. Some of mine are pretty obscure health related areas, while others are pretty massive and mainstream (like bankruptcy law, for example).

If you’re just starting out, I’d say you’re better off picking something product based and kind of obscure. Think “fireplace andirons”, not “LCD TV’s”.

Getting Keywords

Once you have your niche, toss the main term into the Google keyword tool. Tweak the setting to only return keywords with greater than… X searches a month. Whatever your basement is from the math we went over earlier. If you’re literally just starting I’d say ~300 searches. You can also limit the results based on CPC. Once you get your settings down, generate the results and sort them by local search volume. Then I’ve got some bad news… you need to do math. Export those keywords to excel, and trim down your columns so that you have just the keyword, local search volume, and average CPC. (Columns A, B, and C)

In column D, you’ll want… =((B2*.25)*.15)*(C2*.68)

That will take the monthly search volume (cell B2), and assume a 25% CTR, giving you your traffic number when you hit #1. Then it assumes a 15% CTR on your Adsense ads, giving you the number of Adsense clicks. Then it takes the average CPC (cell C2) and multiplies if by 68%, the split you get as a publisher. Multiply these two numbers together, and you’ve got your estimate monthly value for the #1 ranking. Sort by this column. See that top keyword? That’s likely your main target.

At this point, things get a lot less mechanical. You need to look at the competition and gauge how hard it will be to overtake the #1 position. If you’re aiming low, this shouldn’t be hard. If the exact match .com is available, it will likely be a piece of cake. But, that’s a whole other post.

Once you have your main keyword, head on over to Google and search it. Then pull up the Wonder Wheel. See those ~5 terms floating around it? They’re keywords Google expects to see with your primary term. They’re the topics for your other pages. Toss each into the keyword tool and get the highest value permutation from the steps we followed above. You should end up with 5-6 closely related terms.

Next up we’ll cover competition assessment, and getting a domain name. If there’s any questions, fire away in the comments.