How to Maximize Your Profits by Understanding Your Keyword Traffic
Classifying the intent of keyword searches
Every time someone does a search, there is a reason, or intent to the search. What are they trying to accomplish? Are they looking for reviews, general information, to buy a product or simply to satisfy curiosity?
The best place to start is to export all of the keywords in your analytics to a spreadsheet. I am using Google Analytics:
Navigate to Traffic Sources -> Keywords
Make sure that you tell Google to display more than 10 keywords. I typically grab the top 500 for a 6 to 12 month range.

Once exported, it’s time to start sorting the data. At the top of the spreadsheet you will see information related to total visits day by day, week by week, or month by month depending on how you are viewing your data. For this purpose, I deleted all of that information. You should now see a table with the following data:

It creates 10 different header sections:
- Keyword: the keyword that was used to find the site
- Visits: How many visits that keyword brought in
- Pages/Visits: how many average pages were visited when someone found the site under that keyword
- Avg. Time: average time spent on the site
- % New Visits: how many are completely new visits. My experience has said this is highly unreliable data.
- Bounce Rate: percentage of visitors who leave within 8 seconds
- Visits again: this is for your convenience when comparing other pages
- Lead Generated: this is user defined conversions, if you have set up any goals
- Goal Conversion Rate: the conversion rate for your conversions
- Per Visit Goal Value: if your conversion rate is assigned a dollar value, you can see how valuable the visits from each keyword equal.
Now, add another column labeled “Class”. Highlight all the data and click insert -> table. You will now have data that looks something like this (I have hidden some of the columns):

By the way, this is data from an SEO test site and I would prefer to keep it and the keywords used to find it private... sorry for the blurred keywords.
Now we need to classify our keywords. The best way to do this is turn your spreadsheet into a form and insert a drop down box in the class cells.
Enter the different classifications in the columns to the right of the table. Highlight the first cell in Class.
Go to Data -> Data Validation -> Settings -> List

Highlight the columns where your classes are for the source:

Copy the contents of the first cell in Class to the rest of the cells in Class.
You can create your own or use these, but these are some that I have used for different projects successfully.
- General: this keyword is too generalized to understand the searchers intent. For example, if someone searches Iphone, their intent could be many different things.
- Navigational: this search indicates that the searcher is searching for a specific site or a specific part of a site that they have visited before. For example, a search for apple.com would indicate that the searcher is looking for that specific site.
- Transactional: the searcher is indicating that she is ready to buy. For example, “iphones for sale”, “used iphones”, “iphone prices” or “buy iphone”. Even though they may be in the research phase of their searching, it indicates that the end result will be a purchase.
- Geographical: someone is searching for a product, service or information based upon a location. For example, “Apple store Atlanta” or “Iphones tampa”. These types of searches may also indicate intent to buy. However, it is a good idea to understand how many of your sales are local generally.
- Informational: this individual is in it for the information. All they want to do is find out how something works or solve a problem. For example, “the best Iphone apps” or “how to unlock an iphone”.
One note about affiliate offers: When tracking affiliate offers, Google has provided a way to set up conversions when outgoing links are clicked:
Google Analytics provides an easy way to track clicks on links that lead away from your site. Because these links do not lead to a page on your site containing the tracking code, you will need to tag the link itself with the _trackPageview() JavaScript. This piece of JavaScript assigns a pageview to any click on a link - the pageview is attributed to the filename you specify.
Analyzing your data
Once you have classified all of your keywords, you can begin to study your data for trends that will help you improve your conversion rate. Then we filter all of our results to include only keywords with 50 visits and sort them by the highest conversion rate.
It may look something like this:

As you can see from the data above, the keyword with the highest conversion rate is a transactional keyword. So, I know that if I focus my attention on getting traffic for that keyword I will have high conversions.
Here are some more ways to use your data:
- Is your number one converting keyword sitting at position 7 in Google? Can you spend time and resources to move it to the top 3, or even number 1? Some data I have seen suggests that the number one position in a Google gets as much as 67% of the click throughs.
- Up your PPC spend for keywords that are converting well
- Determine which informational searches have low bounce rates, good time spent on site and conversion rates. Spend more time developing these resources and promoting them to gain exposure and inbound links. This is an easy way to develop link bait from existing content.
- Cut your losses. If you have been working your butt off to rank for a certain keyword and it is just not benefiting you, stop. I spent about six months working to rank for “SEO consultant”. I maintained a top five ranking for six months and discovered that keyword phrase brought my business absolutely no value. I did not get one lead!
- Determine what type of searchers you need to target with new content. Other factors will have to include competition and ranking ability. However, you can see what portion of your time should be spent creating informational or transactional content.
Also, keep in mind search volume when making these types of decisions. Even though a keyword may have a high conversion rate, if there is not enough data to say conclusively, don’t bet on it. For example, if you have a keyword with only 10 visits, and one is a conversion making the conversion rate 10%, don’t assume that you will maintain a 10% conversion rate when you have 1,000 visits from that keyword.
This is a guest post by Jason Capshaw, founder of MyWebTronics, an Atlanta Web Design company. He resides in Atlanta with his wife and two children.
- Using Match Types for Negative Keywords with AdWords
- The Beauty of Dynamic Keyword Insertion - DKI
- Benchmarking with Google Analytics
- Top 5 Tips to Increase Conversions
- Google Launches AdWords Remarketing: Its Spin on a Classic Advertising Technique
- Top 5 Ecommerce SEO Tips
- Three Ways to Use PPC to Test SEO Strategy
- Market Research with 5 Great Google Tools


classifying keywords by intention
This is a very interesting article. I wasn't aware of the importance of classifying keywords by intention, but now I can see a practical use of it.
comparative category
One more category of searchers intention could be comparative. Like when someone enters seo vs ppc, or similar.
Great post. Bookmarked for
Great post. Bookmarked for future detailed reading. I don't spent much time analyizing my stats. I should.
Post new comment