Tutorial – Search Cloud (Tagxedo & Google Analytics)

Over the past few months, my WordPress Theme shop has been getting a good deal of organic traffic. After having seen a cool render of one these word clouds for a political issue (using Tagxedo , requires Silverlight), I decided to give it a shot with some of my Google Analytics organic search data, this tutorial will show you how to use both these applications to achieve this final result:

The really cool thing is that there’s a bunch of opportunities to customize the output of such charts, and I’ll show you just how:

google analytics data

The first thing you’ll want to do is login to your GA account, navigate right to your “Keywords” traffic section and export the data like so (choosing “CSV for Excel”, assuming you have Microsoft Excel, otherwise a regular CSV will do):

excel manipulations

In order for Tagxedo to be able to know how much weight to place on each word, you’ll need to provide it with the numbers too. The final format will be a single column in which each cell will have the keyword/keyphrase followed by the total search count for that phrase. The format will also need to be structured exactly like this “wordpress:500“. To achieve this, all we need to do is create a third column whereby we use the “Concatenate” function to bring these values together, like this (A & B are supplied by Google Analytics, we created C):

Once that is complete, you’ll just want to copy the content of the final column. I had tried a bunch of combinations till I saw what actually worked best, but to make a long story short, I found that taking the 100 most frequent keywords was the best approach towards a well balanced word cloud (as opposed to taking the entire column, i.e. 500-1000 keywords/key phrases).

tagxedo customizations

Just before we jump into Tagxedo, I’m first going to create my color palette (and there’s no better tool then Color Scheme Designer 3).

This is where it gets fun (assuming your clipboard is also holding the 100 top phrases including search counts) . Go into Tagxedo and start off by loading up your word list. This will generate a word cloud that usually isn’t that pretty (matter of taste I guess). This is where you’ll want to start customizing heavily, and in the following areas:

  • Shape (although complex shapes will lose the message quite quickly)
  • Fonts (upload them directly from your PC)
  • Color Themes (create one like I did above)

Once you’re happy with the result, go ahead and export a high-res version of the image (otherwise you’ll spend a bunch of time recreating it). At this point, you’re done, quite easy right? Feel free to share your word or search clouds.

SEO KPI: Exposing the Long Tail

This is part of my series on SEO KPI’s (Key Performance Indicators). I found it hard to find this sort of information before (presented in an unbiased manner) and have as such begun compiling these articles. I hope they help you!

What is the Long Tail?

If you’re only interested in the KPI part, skip this section. The term The Long Tail was pioneered by Chris Anderson of WIRED magazine when he first published an article about it in 2004.  To quote him directly:

The theory of the Long Tail is that our culture and economy is increasingly shifting away from a focus on a relatively small number of “hits” (mainstream products and markets) at the head of the demand curve and toward a huge number of niches in the tail.

How does this apply to SEO and your website? Quite simply, the richer and more detailed your content is, the more exposed it will be to an even larger combination of keywords people search on. Imagine for a second you sell clothing online and rely on the images within your e-store to describe the product visually to the client. The typical and popular keywords flowing in could include something like “blue shirt”, but the theory behind the long tail is that there are so many other search terms that hardly get any traffic, but when put together, their numbers become very significant. A typical long tail search for the same scenario could be “men’s navy blue french cuff dress shirt from calvin klein”, but how is anyone ever supposed to find your product if those sort of descriptive keywords aren’t even listed?

Continue reading “SEO KPI: Exposing the Long Tail”

SEO KPI: Breaking New Ground with Non-Branded Keywords

The SEO KPI Series is a collection of posts that aim to enable the reader to extract more meaningful statistics from their analytics package.

Branded vs. Non-Branded

In the first part of my KPI series, I’m going to tell you about measuring what we call “Non-Branded” Keywords. To me, this metric does a great job of understanding “pure” organic traffic. For example, you represent the company “Nike”, the branded keywords would be “Nike air”, “Nike golf”, etc. Your on-page SEO optimization could be rather mediocre and you’d still rank quite well for these terms. But what about keywords such as “tennis gear” or “running shoes” which do not include your unique brand name? The terms are far more generic. There’s a good chance the consumer is more likely to have an open mind when it comes to picking a product (or in this case, brand). Being able to identify this sort of search traffic and build on it has a larger potential of bringing you entirely new clients, not just existing clients who want to see what the latest products of a certain brand look like. The traffic you receive from non-branded keywords is in other words directly proportional to how your brand awareness is performing. Continue reading “SEO KPI: Breaking New Ground with Non-Branded Keywords”