Sponsored Links

Keyword Research Tips for SEO

Researching keywords is one of the fundamental tasks of any SEO efforts. Basically there are a couple of keyword research tools that everyone is using.

Common Keyword Research Tools are:

While these tools produce a nice list of keywords and their respective number of searches, I have to warn you - the number of searches per keyword is very inaccurate.

The data on the number of searches is critically important. You can work on keywords for months and when you finally clinch the top spot you may discover that the number of daily searches per keyphrase is about 100 to 1000 times less than the above tools predicted. I will give you examples below.

The number of searches per keyword is inaccurate because:

  • Overture and Wordtracker aggregate keyword variants and phrases (single, plural, misspellings, bigger phrases into smaller phrases etc.)
  • Wordtracker uses data from meta search engines that have different demographics than the major search engines

Never ever believe the Wordtracker and Overture number of searches per keyword. I've learned this the hard way.

There is a much better and completely free way to estimate the number of searches per keyphrase on Google! I have found it to be much more accurate than Wordtracker and Overture. When you have the rough number of searches on Google, you can (in 99.99% of the cases) safely assume than Yahoo and MSN will provide less traffic than Google for these keywords.

How to estimate the real traffic per keyphrase (number of searches x CTR)

The answer is simple - Google Adwords Traffic Estimator

Here is the procedure. Let's say you want to estimate the number of searches per day for the keyphrase "calorie counter". At the moment I write this newsletter, Overture and Wordtracker provide this data (using DigitalPoint's keyword suggestion tool):


"calorie counter" on Overture -3587 searches per day
"calorie counter" on Wordtracker - 3072 searches per day

This seems like a very high-volume keyword. But is it?

Checking Google's AdWords Traffic Estimator.

  1. Go to Google AdWords Traffic Estimator
  2. Put [calorie counter] in there. Using the [] to enclose the query means you want to bid when someone searches exactly for the keyphrase calorie counter (that will exclude searches like activity calorie counter, food calorie counter etc.)
  3. Select the various options to make the campaign such as Global searches or Regional etc.
  4. Select the countries you wish to estimate searches from. "All Countries" would work great for estimating global number of searches
  5. Now set the Maximum Cost Per Click to $100. That will put you in position number #1 which gets the highest click-through ratio.
  6. Now we get an estimate of 94 clicks per day. Compare 94 to 4000? That's a huge difference. The traffic estimate is based on the CTR of the ad placed on the top spot for the given keywords on Google and their search network.

So what is the real number of searches for the keyphrase calorie counter on Google? It is roughly 94 / (divided) Google's estimated CTR for the top ad spot on these keywords.

The difficult part here is to guess the estimated CTR that Google uses. We also have to factor in that the estimated number of clicks includes searches on other search engines and that it includes second, third etc. page clicks (Google shows the given ad on second, third etc. SERP pages).

I use these coefficients to estimate the real traffic (not number of searches) potential of keywords on Google. It is based on my experience with the keywords I have monitored.

To find the traffic potential of the given keyphrase (in our case, calorie counter) on Google multiply the number of clicks (94) by:

  1. 5 to 10 for SERP position #1
  2. 2 to 5 for SERP position #2
  3. 1 to 3 for SERP position #3

For calorie counter we get these estimates:

  1. 470 to 940 clicks per day being #1 on Google for calorie counter
  2. 188 to 470 clicks per day being #2 on Google for calorie counter
  3. 94 to 282 clicks per day being #3 on Google for calorie counter

The above ranges for the top 3 SERP spots will vary from keyword to keyword. As I've written, I have based them on the queries I monitor. The accuracy will depend on the CTR of the top spot ad on AdWords.

My experience has shown that the top 3 spots in the SERPs will in most cases get at least 10 times less traffic than Overture and Wordtracker predict. Being ranked below top 3 will have a more drastic difference.

Here's another example. Consider the keyphrase big muscle.

  • "big muscle" on Overture -1132 searches per day
  • "big muscle" on Wordtracker - 772 searches per day
  • [big muscle] on Google - 4.1 clicks per day

My Traffic Estimates on Google's SERPs (for big muscle)

  • Position #1 - 20 to 41 clicks per day
  • Position #2 - 8 to 20 clicks per day
  • Position #3 - 4 to 12 clicks per day

My estimates may or may not be super accurate, but I have found them to be generally accurate in most cases. Compared to Overture and Wordtracker they will be much more accurate. In my experience, the volume of searches per most keyphrases on Yahoo and MSN is considerably lower than on Google at the moment. Google is still the most important search engine in terms of traffic potential.

If you have top positions for some keywords and track them, run the above calculations on Overture, Wordtracker and Google AdWords Traffic Estimator data and let me know on the forum how accurate is the procedure I have proposed in this newsletter.

If you don't have top 10 rankings and want to spend money on Google AdWords you can find out a better estimate for the number of searches by running an AdWords campaign where:

  • you bid enough to be in top 8 (first page ads) for your keywords
  • you bid on [your keywords]
  • you bid only on Google's search engine

The number of impressions produced by such a campaign will indicate the real volume of searches, but this method costs money. Although it includes impressions on second, third etc. SERP pages, most searchers don't go beyond the first page of results and you can assume this data to be fairly precise.


Latest SEO Blog Entries

Shareware Marketing 101 - 28 August 2006
Google Webmaster Central To Solve Canonical Issues - 16 August 2006
Forum Upgraded With Signatures and Avatars - 13 July 2006
Climbing the Keyword Ladder - 17 June 2006
PayPal Is Not Enough - 12 June 2006
SEO Guide Gets a New White-Grey-Black Hat Skin - 06 June 2006
Matt Cutts On BigDaddy, PageRank and The SandBox - 17 May 2006
Google Patents On PageRank Variants - 11 April 2006
Microsoft Paper Gives Clues Into The Future of SEO - 10 April 2006
Focus On The User And Get More Traffic And Revenue - 09 March 2006