How do you make short tail keyword determinations when combined with long tail when there is not enough search volume to provide info

[irvingw]

Confusing question, allow me to elaborate

We have a few pages that target a particular doctor for example. One of those pages is about his backround.

His short tail is his name "Dr Irving Weiss" for example, low competition of course and already too low of search volume to show in Google keywords tool (which i know isn't the best tool)

so now one of his tab pages addresses info on his medical background (credentials, schools, awards, sanctions)

if you search for those single keywords alone you get something like this

hospitals 74,000

background 673,000

credentials 49,000

but that doesn't necessarily mean more people will search like

"dr irving weiss background"

than they will

"dr irving weiss credentials" just because background has more searches

"dr irving weiss background" and "dr irving weiss credentials are way too low search volume to have any data on, so how can you come to a proper keyword targeting conclusion when the data is not there?

THANKS IN ADVANCE for any insight!

[irvingw]

thanks, i'm going with a combination of two.

one seems to be decent and also intuition tells me it's good, and other one has no other sites currently well optimized for it, so it actually MIGHT bring more traffic than the better one because it's an easy win with less competition.

[Vizergy]

Absolutely.  It just seems that at the end of the day it's going to be a guess so the most data you can put towards it the better your chance are of choosing correctly, of course.  I just hope that when you do make your decision it ends up being correct for at least the majority of the pages.  Good luck - I have faith in ya!

[irvingw]

This is so incredibly long tail - we're looking for a combination of the doctors actual name which is already long tail. Not just "Dr keyword", so it might not be that obvious.

Thanks for the help thinking this through.

[Vizergy]

Google Trends shows 'Dr Credentials' as the one to go with...

[Vizergy]

"this is a one time change that would affect about 800,000 pages so we want to get it right the first time around and stick with that url."

Oh yeah - I dig that.

Maybe instead of just the one word checks simply make it more broad?  Like 'Dr credentials' and 'doctor background check"?  Those would probably register and would be a bit more relevant to your campaign... maybe?

[irvingw]

thanks billy,

this is a one time change that would affect about 800,000 pages so we want to get it right the first time around and stick with that url.

my instinct would be to pick the best two and combine them in the url, then optimize for the best one out of those as primary target, and create a few instances of the secondary in the content as well in an attempt to capture both.

The question is which is the primary, do we go with search volume for that single broad keyword or use intuition which is quite often wrong hmmmmm stumped a bit

doctorsite.com/dr-irving-weiss/background-and-credentials

<title>Dr Irving Weiss Background Check | Credentials and Hospital Affiliations</title>

[Vizergy]

Mr. Weiss,

I suppose, for me, it would depend on the campaign.  If it was going to be a long running client or if it was going to be my company's website I would pick one phrase and run with it for a few months and then change it up and run the other for a few months and compare the data to determine which one was the most successful.

If, however, it was a short running client like a 'one time' optimization (which I hate but they do happen) I would first rule out that I could in no way run them both together and if I couldn't and if I determined that they both carried equal relevancy to the page's subject I would take the one or two words that make the short tail phrases different (in this case 'background' and 'credentials') and I would find out which of those carried the most search volume and make an educated guess from there - I know you touched on that and I agree that it doesn't 'mean' much - but it does help with getting an idea, in my opinion.

You pose an interesting question and as I have taken your advice many times - and you certainly give excellent advice - I am quite curious as to how YOU or others here would approach this.  I will be watching this thread closely.

Cheers, my friend