Picking the Best Keywords for Your Geotargeted Pages
Since the days of creating my first geotargeted site for Raleigh, NC I’ve struggled with the best sequence of (or ) combination to target. For instance, which of these combinations would capture the largest number of visitors?
• Raleigh Real Estate
• Raleigh NC Real Estate
• Real Estate in Raleigh
• Real Estate in Raleigh, NC
• Raleigh North Carolina Real Estate
Tough call. A few years ago I took a guess and went with a combination of two keywords and tackled it like this:
Raleigh Real Estate in Raleigh, NC.
Well I think that looks a tad spammy so I want to find a better practice that can achieve the same results. So I have been polling family and friends by asking things like “If you were new to the area how would you search in Google to find a place like 518 West?” (Its a local Italian restaurant). Of course my sample size is way small but the response I’ve been getting are all over the board:
• Raleigh Italian Restaurant
• Italian Food Raleigh
• Italian Restaurant in Raleigh
and on, and on, and on. There seemed to be no consensus. So my advice for the best approach is to create multiple pages if possible targetting different variations. If you have enough trust rank and inbound links with varying anchor text you can probably nail two variations on one page.
I also did a few high profile queries like Chicago Accountant and Accountant in Chicago and it seems that the SEOs out there are targetting and nailing more than one term by mixing up the inbound links and displaying several combinations on the page. You’ll also begin to find that Google is getting good at the LSI thing and they can relate North Carolina to NC as seen in this query for Raleigh North Carolina Real Estate where my raleighlist site shows up in the middle of the SERP’s even though the page has no mention of ‘North Carolina’.
If you are SEOing for local geotargetted pages how do you go about it?
Oh yes! I forgot to mention the Adwords trick. This is really the best approach to finding out the best combos.
I like the black-hat method. Launch a site using black hat content generation with every keyword combo you can think of. Once all your pages get indexed, wait 3-4 days. Delete whatever doesnt stick (301 to home page), and replace what does with good content.
We actually have gotten away with this on white hat sites as well. Do some CG, wait for it to get indexed, and then delete the generated content and replace it with good on the pages that stick.
Russ,
That doesn’t tell you anything about what people search for.
I think you have to target different keyword “sets” for terms related/similar to “raleigh real estate” and build pages accordingly (as you mentioned).
I also like to look at total numbers of pages indexed Google and Yahoo for those diferent terms, like:
1) “real estate in Raleigh”
2) “raleigh real estate”
If one term, like #1 noted above had 31,800 results while #2 had 312,000, then we build pages according to total # of indexed competitor pages, often going for the lessor marketed terms.
If a term had 2,500,000 results, and let’s say it was “raleigh real estate,” then we would opt to get the “change left on the floor,” so to speak, by using another keywords with often 3, 4, or 5 keyword phrase variances. Many will argue with this but I really believe that you get higher conversions/leads that way.
Don’t get me wrong, I love ranking for the much more popular terms but let’s get realistic. If 2.5 mil results are already there and you’re just starting out with new pages or a new site, it might take you a year to get there (white hat method) going for common phrases.
Dario’s method above is a realy good one as well.
You’re right, it doesnt say just what people are searching for, it does everything for you. You find out which terms have the best combination of weak competition, impression rate (searches), and click throughs.
It worked fantastic for a personal client of mine (a barter for a wedding reception). While I can’t tell you what exactly was searched most for, I can tell you that the performing keywords were those where locale was second when doing searches involving the word “home” but locale first when doing seaches involving the words “real estate” or realty. ie: “homes city, st” performed best as did, “city, st realty”.
Also, I prefer using the number of results in an allinanchor:”words go here” query rather than simply a “words go here” query to determine competition. I find this tends to give me a better idea of how many people, relatively, are “competing” for that phrase.
I live at 28640 Commonwealth in Seattle. Been up here before?