Keyword Research and Competition Strength WAR

[Kennyz]

Confusing thing #1 - Keyword Research
I started to get confused, today i did keyword research for few hours and i came up with something strange.

How you choose the keyword? Confusing thing is:
If I'm looking on BROAD match the result is: (Example)
Keyword 1 : dog - 49,500 of searches
Keyword 2 : cat - 33,100 of searches

If I'm looking on [EXACT] match the result is:
Keyword 1 : dog - 6,600 of searches
Keyword 2 : cat - 9,900 of searches

Bah, the first question, which one better to choose? keyword #2 have more searches with exact match, what do you say guys - how do you choose your keywords?

Confusing thing #2 - Competition Strength
We will continue with same keywords. now I'm going to check the competition strength to see which keyword to choose right?

If I'm looking on BROAD match competition the result is:
Keyword 1 : dog - 101,000 websites
Keyword 2 : cat - 1,820,000 websites

If I'm looking on "PHRASE" match competition the result is:
Keyword 1 : dog - 112,000 websites
Keyword 2 : cat - 80,300 websites

wow, the numbers is going crazy.

Summary
So now we have the data, how we will choose which keyword is better?
Lets look on the numbers, and calculate the KEI :

Lets take BROAD match searches and BROAD competition :
Keyword 1 : dog
Searches: 49,500
competition: 101,000
KEI: 24,259

Keyword 2 : cat
Searches: 33,100
competition: 1,820,000
KEI: 601

Conclusion: Keyword 1 is the BIG winner, x39 times better then Keyword 2

Lets take [EXACT] match searches and "PHRASE" competition :
Keyword 1 : dog
Searches: 6,600
competition: 112,000
KEI: 388

Keyword 2 : cat
Searches: 9.900
competition: 80,300
KEI: 1220

Conclusion: Keyword 2 is the BIG winner, x3 times better then Keyword 1

So, what do you say about this confusin keyword & competition strength research, which keyword is the winner, and how do YOU decide who wins?

[imfusa]

It would be very tempting for me to go for keyword "cat".
But i just saw the competition so i would sacrifice 3300 searches for almost 20 times smaller competition.

[WinstonT]

Well, from this example, we won't know for sure. Although keyword research deals with a lot of numbers, we often forget that competition is secondary to conversions. Remember... conversion is KING, not competition. A large search number might be great, but since both are pretty similar, I guess it boils down to its conversion. Don't forget that too low competition also might mean that it isn't very profitable in the LR/SR.

If you think a keyword has a hungrier market than the other (by means of spyfu, commercial intent, buying words etc), then by all means, go for that one. If both are as hungry, look at the long term aspects. Is it a general niche, where by the broad matches can allow targeting of other products? Commissions? Can I build a list from there? Any future moves?

I'll admit I can't find an answer to this question! Oh, infuriating.

[thecableguy]

JMO but the number of competitors isn't the best way to determine the level of competition. It doesn't matter if it's only a thousand or one million, page one of the SERPs is what matters, the searches drop significantly for subsequent pages. I don't even try to filter potential keywords in MarketSamurai anymore, instead I just go strait to the SEO checker module.

[Kennyz]

Originally Posted by thecableguy

JMO but the number of competitors isn't the best way to determine the level of competition. It doesn't matter if it's only a thousand or one million, page one of the SERPs is what matters, the searches drop significantly for subsequent pages. I don't even try to filter potential keywords in MarketSamurai anymore, instead I just go strait to the SEO checker module.

I see, but you choose your keyword by EXACT or BROAD searches? that's the question

[NastyBlast]

In my experience it is best to take a bottom-up/inverted pyramid approach. so that would mean you would start with broad.

First get a keyword plan down it's not just about picking a few terms and using only those. If you're in it for the long run you can get a huge compression factor by using the inverted pyramid approach.

identify the high-volume keywords that you believe will convert for you the best. Usually these are 2–3 word terms. Then extend your keyword list to about 100 sub terms that are longtail 5–7 keywords.

deploy your longtail terms... And then slowly work your way up the list in descending order from most words to least words. This is where the compression factor occurs and can ultimately get you a solid top position in natural search listings. With each longtail term you take a position on you are slowly compressing your way to the primary 2-3 word terms.

It is exactly as I described... An inverted pyramid in which your longtail terms are at the base... And because they all have the same primary keywords making them relevant to each other... they will compress to give you a big boost on pages where you specifically target your primary 2-3 word terms. I like to refer to it as an inverted pyramid because of the crushing force that occurs as your longtail terms add up.

with all that weight of the base longtail terms it makes it easier for you to take the smaller primary terms.

In the end you will have achieved a large support structure of pages that give heavy boost to your primary term pages. I know this does not directly address the question you have asked what I'm saying is look at it from this point of view and in the long run you will be in a much better position.

don't get too distracted by volume and competition... A high-volume term doesn't necessarily lead to sales. You will have to figure that part out as you go unless you have worked with this niche before