Search Engine Watch
SEO News

Go Back   Search Engine Watch Forums > Search Engines & Directories > Google > Google Web Search
FAQ Members List Calendar Forum Search Today's Posts Mark Forums Read

Reply
 
Thread Tools
Old 07-12-2006   #1
xxx
Member
 
Join Date: Jun 2005
Posts: 17
xxx will become famous soon enough
Measuring keyword competition

Hi everyone

I am currently working on a project and I am at the stage where I need to select the most effective keywords/keyphrases for my site. I want to avoid highly competitive keywords but don't know How to clearly measure it. Is 1 million website competing for a word/pharse too many or may be 2 million? Is there a recommended threshold? Bearing in mind I have currently 0 page rank.

I would be most grateful if anyone could help
xxx is offline   Reply With Quote
Old 07-12-2006   #2
JohnW
 
JohnW's Avatar
 
Join Date: Jun 2004
Location: Virginia Beach, VA.
Posts: 976
JohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud of
The number of pages competing for a keyword has absolutely nothing to do with how competitive it is.

Assuming your code and content are done correctly, the main driver of how competitive is a particular keyword, will be related to links, i.e., how many/how good/how optimized are the links that point to the pages that currently occupy the top spots for that keyword.

For example if the top results for a particular kw have low link counts and the links come from crummy pages and the anchors are all stuff like "click here" then there is practically no competition. The opposite of that is of course where the fun starts ;-)
JohnW is offline   Reply With Quote
Old 07-13-2006   #3
xxx
Member
 
Join Date: Jun 2005
Posts: 17
xxx will become famous soon enough
keyword competition

Makes sense. Thanks a lot john
xxx is offline   Reply With Quote
Old 07-13-2006   #4
scrubs
UK
 
Join Date: Mar 2006
Location: UK
Posts: 169
scrubs is on a distinguished road
Quote:
Originally Posted by JohnW
The number of pages competing for a keyword has absolutely nothing to do with how competitive it is.
I would tend to disagree with this comment, sorry JohnW If I want to compete in the finance market for 'credit card' for example, I'd look at the results shown and number of pages I would think that is really really competitive.

For nicher markets, then yes the number of pages thrown up is not a indicators as it is the other end of the scale.

The results in the SERPs are the best indicator of how competitive a search is.
scrubs is offline   Reply With Quote
Old 07-13-2006   #5
JohnW
 
JohnW's Avatar
 
Join Date: Jun 2004
Location: Virginia Beach, VA.
Posts: 976
JohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud of
>I would tend to disagree

You can certainly disagree but that does not change the facts. The number of results returned for a keyword has nothing to do with how difficult it is to get a site ranked for that particular keyword. The level of competition has everything to do with it.

The top page will be at the top of the pile regardless of how large the pile is. Top pages are defined by an algorithm that considers many factors however the *quantity* of competing pages is not one of the factors considered nor should it be.
JohnW is offline   Reply With Quote
Old 07-13-2006   #6
scrubs
UK
 
Join Date: Mar 2006
Location: UK
Posts: 169
scrubs is on a distinguished road
OK, fair enough.

Question though...as a fellow SEO'er would you be happier to achieve a page 1 listing out for 100,000 results or 100,000,000 results?

I'm sure you would use the result in the 100 million results in a case study?
scrubs is offline   Reply With Quote
Old 07-13-2006   #7
Chris Boggs
 
Chris Boggs's Avatar
 
Join Date: Aug 2004
Location: Near Cleveland, OH
Posts: 1,722
Chris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud of
I think I would have to say that neither of those options would necessarily look "better" in a case study, especially if you are presenting it to savvy viewers.

I have to agree that competitiveness is only measured by the competition in the top 30, let alone the top ten. One way I have used in the past to target highly competitive keywords is to look for the "Achilles’ heel" within the top ten and go after that specific site. Once you can supplant it, then you are on the way to competing against the "tougher" ones.

What determines "tough" competition in my book? The usual suspects: linking, percentage of deep links, saturation, unique content, traffic, etc...

Last edited by Chris Boggs : 07-13-2006 at 09:51 AM.
Chris Boggs is offline   Reply With Quote
Old 07-13-2006   #8
scrubs
UK
 
Join Date: Mar 2006
Location: UK
Posts: 169
scrubs is on a distinguished road
Good points Chris.

My last post regarding page 1 listings was purely based on a good search term, something that was of value to the client.
scrubs is offline   Reply With Quote
Old 07-13-2006   #9
JohnW
 
JohnW's Avatar
 
Join Date: Jun 2004
Location: Virginia Beach, VA.
Posts: 976
JohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud of
>would you be happier to achieve a page 1 listing out for 100,000 results or 100,000,000 results?

I couldn’t care less one way or the other.

>I'm sure you would use the result in the 100 million results in a case study?

If you were to include this type of irrelevant data in a sales context it would be unethical and misleading. I just wouldn’t do it. What matters are the results, i.e. top rankings for keywords that are important to the prospects business plan.
JohnW is offline   Reply With Quote
Old 07-13-2006   #10
scrubs
UK
 
Join Date: Mar 2006
Location: UK
Posts: 169
scrubs is on a distinguished road
Quote:
Originally Posted by JohnW
What matters are the results, i.e. top rankings for keywords that are important to the prospects business plan.
I totally agree and this is my work ethic.
scrubs is offline   Reply With Quote
Old 07-13-2006   #11
Anonymous
Sneaky Web Development
 
Join Date: Jul 2005
Posts: 13
Anonymous is on a distinguished road
But, most often, it may be safe to assume that the more there are sites for a certain keyword, the more competition, the tougher the competition is.

Unless all the 100 bil websites are not competing and noone really wants to be the best there.
Anonymous is offline   Reply With Quote
Old 07-13-2006   #12
Komodo Tale
Member
 
Join Date: Mar 2006
Location: Seattle, WA
Posts: 17
Komodo Tale will become famous soon enough
First off, good rankings mean nothing unless the people who are looking for the service or product you offer are using that keyword query. I'd rather rank higher for a less common query if that query brings-in highly targeted prospects.

As for measuring keyword difficulty, tools like Wordtracker and Overture can give you simple baseline numbers, but you then have to look at the websites in the SERPS and compare the content, coding, and inbound links for the top ranking websites against known website ranking factors (http://www.seomoz.org/articles/searc...ng-factors.php) to understand why they rank where they do.

To measure best practices I look for websites that rank well for several related keywords or phrases then statistically determine how their ranking factors stack-up as a community.
Komodo Tale is offline   Reply With Quote
Old 07-17-2006   #13
Black_Knight
Ancient SEO
 
Join Date: Jun 2004
Location: UK
Posts: 152
Black_Knight is just really niceBlack_Knight is just really niceBlack_Knight is just really niceBlack_Knight is just really niceBlack_Knight is just really nice
I use a metric which is based on taking the number of results for the phrase when searched on Google, then using the top Overture bid for the term as a multiplier.

The more expensive the keyword in PPC, the more the chance that some serious professional SEO is used in the SERPs, as opposed to it merely being a popular term appearing in millions of pages, but that noone actively optimised for.

<Added>I have to agree with John, of course. There are, according to Google, 28,300,000 pages for the word Cialis, and 85,200,000 for bedding. If you really think that running for a top rank for bedding is more than three times harder than ranking #1 for Cialis you had really best think about a different career. </Added>

Last edited by Black_Knight : 07-17-2006 at 06:54 PM.
Black_Knight is offline   Reply With Quote
Old 07-18-2006   #14
Chris Boggs
 
Chris Boggs's Avatar
 
Join Date: Aug 2004
Location: Near Cleveland, OH
Posts: 1,722
Chris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud of
Quote:
Originally Posted by Black_Knight
<Added>I have to agree with John, of course. There are, according to Google, 28,300,000 pages for the word Cialis, and 85,200,000 for bedding. If you really think that running for a top rank for bedding is more than three times harder than ranking #1 for Cialis you had really best think about a different career. </Added>
Greetings! I'll have to argue this one with you, since I think you are actually inferring that it would be harder to rank for “Cialis” than for “Bedding.” Cialis is a branded term, and bedding is about as general and bland of a term you can find. I would venture to say that the number of websites actually competing (actively) for terms with "bedding" included is likely to be more than 3 times the number of sites competing for Cialis. But I could be wrong…

Cialis, due to its unfortunate additional scrutiny associated with being in the "sneaky" pharma industry, may gain some degree of difficulty points. Conversely, if the site is the official Ely Lilly product site, it may get a veritable “free pass.” My top ten for this term at Google includes the brand site (side note: too bad they haven’t tagged their site for Co-op yet in order to appear in the refined results, eh?) as well as three .gov TLD sites, and 6 sites that can clearly be argued to be commercial in nature. Judging from the forums in there, this very thread has a shot. Still probably very difficult to get in the top ten, especially the top 4. Also, sometimes an indicator of how many sites may be competing SEO-wise is the number of sponsored listings, which is 29 for "Cialis" (by my count at my time of search)(this is similar to your Overture multiplier,BlackKnight).

“Bedding,” by my search, has some surprising results. LNT.com has done a good job of optimizing, and Amazon has the added weight of ranking in search results forever. Of course all the top ten have major inbound links. Probably tough to get listed here, especially starting from scratch. There are just over 200 sites (click on "next page" a few times...) competing for this term in sponsored listings, though. WOW! I would say many more people may be looking for organic optimization especially in that tough market…

So I'd say it might be twice as hard to rank for "bedding" as for "Cialis." Of course I would still want to study the top 50 or so across the big three before estimating the difference in difficulty between these words.

Good example and maybe some others will throw their initial (or researched) thoughts into the thread!

Last edited by Chris Boggs : 07-18-2006 at 09:54 AM. Reason: added links to searches
Chris Boggs is offline   Reply With Quote
Old 07-18-2006   #15
Black_Knight
Ancient SEO
 
Join Date: Jun 2004
Location: UK
Posts: 152
Black_Knight is just really niceBlack_Knight is just really niceBlack_Knight is just really niceBlack_Knight is just really niceBlack_Knight is just really nice
Chris, good points. However, quite recently I was hired by a manufacturer/distributor of a product line to run a reasonable advanced SEO workshop for all his affiliates and retailers. I told them all to get into AdWords as a very minimum level of investment. There were over 80 people at that single seminar.

So Eighty bidders for a term only means that that industry has some degree of interest in SEM. That could be because a manufacturer sent out a few tips, or because a locality specialises in that industry, or because an SEO/SEM firm specialises in that niche and actively hunts out such dealers.

I always go for the bid prices, because at 10c per click, there's probably no real incentive to do more. At $10 per click, the company just has to consider that even a top-flight expensive quality SEO consultant or company is still good ROI. The higher the bid price, the higher the chance that the company can also afford serious, industry grade SEO, not just let their IT guy having a go at it.
Black_Knight is offline   Reply With Quote
Old 07-19-2006   #16
Chris Boggs
 
Chris Boggs's Avatar
 
Join Date: Aug 2004
Location: Near Cleveland, OH
Posts: 1,722
Chris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud of
Thanks Black Knight. That makes sense. You could also add the number of people bidding over a certain thresholds as a multiplier of "serious" competition. Wondering now what the bids are for Cialis and bedding, although I would prefer to compare apples-to-apples...

Now that I think of it, your argument fails to consider the super giants out there that are going to do SEO and continue to find ways to bid less on terms, no? Amazon would be an example of someone who probably just "threw bedding in" one day and was ranked top 10 within a month. That is probably not a great example since they rank #3 at G but not to be found at Y! They do not bid on "bedding," but I would be pretty sure they are bidding on some longer tail terms with lower CPC...

In the ongoing case of Cialis vs bedding, Bedding has an Yahoo estimate of 120030 searches in June (summer nights? Lots of people moving?) and has a log jam at the top 3 bids at .77,.76,.76. With a (very) hypothetical 10% CTR for #1, we are looking at close to $10k spent per month using the Y! number (note: This is a Yahoo! environment test only. The WordTracker "Overall" searches numbers were pretty close to the Yahoo estimate, so IMO the Y! estimate is a little high due to inclusion of automated clicks, etc...However if I included this I would have to estimate click costs at Google, etc...Leave that to the software).

Cialis has a higher bidder at Y, bidding 1.31 (paying 1.24). #2 is, you guessed it, 1.23, but only paying .91. Drops from there. Same 10% CTR would lead at Y! Est. June searches (101863 - summer nights? Lots of people moving? ) to exposure of closer to $13K, or 30% higher.

Based on just the top bids, you would then say that it is 30% harder to rank for Cialis? How "deep" do you go? You could say it's even harder, I guess, since #2 and #3 both have greater exposures, CTR's considered equal...

I like the idea, but I still think that the focus should be more on the top 30 or even 50 of the organic rankings. A simple test of presence of SEO factors can yield much more information about the field of battle, in the long run, than a PPC snapshot can. But as I said, I do like the PPC "litmus test."
Chris Boggs is offline   Reply With Quote
Old 07-19-2006   #17
JohnW
 
JohnW's Avatar
 
Join Date: Jun 2004
Location: Virginia Beach, VA.
Posts: 976
JohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud of
I think we are going down a rabbit trail here but OK.

Chris – I understand what you mean about looking at competition in the top-30 but it’s very also interesting sometimes to see the onpage/offpage factors of some of the better sites that are in positions 30-200 and compare them to the ones in top-30 as to why they are not in the top-30.

Back to rating the competitiveness of a keyword - I think it is legitimate to say that, as mentioned, the number of searches, avg. PPC bids, number of bidders etc. are all *indicators* of what you might find when you do a proper competition analysis …. but at the risk of sounding like a recording of myself – these things are not part of the algo. It’s just noise. The fundamentals of content, code and links are what matters, and of course links are the wildcard.

When I do a comp. analysis I always assume that when I am done with the coding/content the site will be as good or better than every competitor. At that point I remove onpage thinking from the equation, and look at links. IMO link quality, quantity and anchors are what matters at that point.
JohnW is offline   Reply With Quote
Old 07-19-2006   #18
Chris Boggs
 
Chris Boggs's Avatar
 
Join Date: Aug 2004
Location: Near Cleveland, OH
Posts: 1,722
Chris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud ofChris Boggs has much to be proud of
Sure John...it is a bit "noise" but sometimes there are hidden messages in the noise...and it's "fun" noise to tinker with, kind of like those math tricks I have long forgotten that give you a fairly accurate (gu)estimate with 1/10th the work. Either way, thanks to everyone for keeping this thread interesting and probably helpful to those wrestling with kw research.
Chris Boggs is offline   Reply With Quote
Old 07-21-2006   #19
DONMAC
Member
 
Join Date: Jun 2006
Posts: 6
DONMAC is on a distinguished road
This commentary proves...

All this back and forth commentary proves one thing to me:

SEO is as imprecise an endeavor as I ever would want to get involved in for my ad agency clients. Humbug.
DONMAC is offline   Reply With Quote
Old 07-22-2006   #20
JohnW
 
JohnW's Avatar
 
Join Date: Jun 2004
Location: Virginia Beach, VA.
Posts: 976
JohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud ofJohnW has much to be proud of
>Sure John...it is a bit "noise" but sometimes there are hidden messages in the noise...and it's "fun" noise to tinker with

Ouch… I feel a bruise forming. Chris, I truly didn’t mean to belittle the discussion - as you say, it’s good to tinker and learn. I guess I am overly sensitive. As a professional SEO I talk to people everyday who have been grossly misinformed about stuff like this by low-end competitors, it is driving me crazy. It may have already done so ;-)
JohnW is offline   Reply With Quote
Reply


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off