Picking keywords - balance Google search volume and competition

[JavaReb]

SEO newbie here. Ive been reading everything I can get my hands on about the subject, watching videos etc.
Im trying to get a handle on the data im looking at while using the keyword tool. Ive started with a list of a few "starter" keyword/phrases and now im trying to choose some of the keywords from the list. Im trying to balance the competition vs search volume. That part I have down ok I think. Now, what about viewing the competing number of pages? If you click on the keyword, you will get a SERP (search page) with the results based on that keyword. If you look closely, you will see the # of results that the search returned. Lower is better, correct?

Example.
some key phrase
competition = medium
(global)search volume = 5400

However, if you click on the keyword, you get over 12 million results. Should I write off this keyword, if I have to compete with this many other pages?

This is all in regards to writing article/review content on my website.

ideas/comments?

[aristotle]

Those total results numbers are often inaccurate and don't mean much anyway.

If you create a good website, your ultimate competitors will be the pages in the top ten, so those are the ones you need to study. If the top positions are occupied by sites like Wikipedia, Amazon, Youtube, and/or well-knowm brand names, then it will be difficult for you to get to the top. If weaker sites are in the top ositions, then you will have a better chance.

[tedster]

There's quite an art to using Google's keyword data. For instance, you may be seeing SEARCH volumes (however dependable) but that is not the same as knowing the CLICK volume. Especially for short phrases and single words, as many as 40% of the searches can result in no click at all.

Then there's the highly questionable "number of results" that is reported. Those pages may or may not contain the search words, but they're certainly not all competitive. The number of Adwords running on the SERP gives a better sense of the level of competition.

The new external keyword tool also shows a bar graph that indicates Google's idea of the competition level.

[aakk9999]

You may want to see how many results you get with operators allintitle and allinurl for your keyword phrase - this may give you little more indication on how many competitors are likely to be targeting that particular phrase.

[JavaReb]

aakk9999 - yes, ive been doing that too.

For example
keyword A - search vol = 720, 1,070,000 results, 658 allintitle, CPC = 2.67
keyword B - search vol = 3600, 135,000 results, 8900 allintitle, CPC 1.83

[buckworks]

as many as 40% of the searches can result in no click at all

A well-writen description that attracts more clicks than the other guy can be a big competitive advantage. Craft your page descriptions carefully, and if the search engines are displaying snippets from the page, experiment with your on-page text to see if you can make the snippet more appealing.

JavaReb, the type of research you're describing is useful for general ideas to get started, but remember the limitations and do not stress about the exact numbers. I agree with Aristotle that it actually matters very little how many total results there are for a certain search, what matters is how smart the top few dozen competitors are and how aggressively they're targeting the phrase you want.

In some searches, the top results are there mostly by luck, not because they've made a point of taking aim at the phrase. In others, the top results have worked long and hard to get there. It takes human judgement to discern the difference, and competitive reality as seen through those lenses can sometimes be very different from what the keyword competition number crunching might suggest.

Another important question to ask is about the user mindset ... which keyword would be more effective at connecting with users who were ready (or nearly ready) to do the action that would put money in your pocket?

Example: If Keyword A attracts users who are looking for ideas to customize the widget they've already purchased, that might not be a very useful target for you as a widget seller. Keyword B might to do a much better job of reaching users who haven't purchased their widget yet.

Food for thought ...