Archive for May, 2009

May 11
2009

Google just announced four new features in Google Analytics. These features are in beta and are being rolled out to all GA accounts so hopefully you’ll have access to them very soon. Two of these features are intended for deep-dive analysis and offer an incredible amount of insight right at your fingertips! If you are interested in saving time and doing better analysis, keep reading. :) The two new features are:

  • Secondary dimensions
  • Pivoting

To find out more, keep reading or watch our video:

Secondary Dimensions

Personally I’ve found this new feature to be extremely helpful. It has helped me focus more on analysis and less on digging through reports (yay!) and it definitely decreased the steps taken to get to a particular report. Secondary Dimensions allow users to view two different dimensions within the same GA report. This makes analyzing your data more efficient and saves you time. Instead of having to run different reports and compare the data, you’re able to run the report and see the data side by side. Let me show you an example:

One of our clients observed a sudden spike in their direct traffic. We needed to ascertain where the traffic was coming from. Since the client had attended a couple of recent trade-shows, our initial assumption was that this spike in direct traffic resulted from the buzz around the shows. Stop – do not settle on this conclusion so easily! We’ve been trained to use data to validate assumptions and conclusions.

In the “All Traffic Sources” report, I selected traffic sources by medium, and then I added a secondary dimension for “Country/Territory”, and voilà, the report was created and it showed us that out of 1972 direct visits, 1037 were from Pakistan.

Google Analytics secondary dimension

Wait a second, we knew that the trade-shows where in the US and not in Pakistan, and the client’s target audience is US-based as well. It turns out that this particular client has an offshore software development office in Pakistan. which explained the recent spike in traffic as the developers were making updates to the site.

Even without the Secondary Dimension feature this same information is available, but you would have to leave the current report and go to a “direct segment” and then look at a geography report to find the information that is now available using the secondary dimensions feature (with one click). As stated earlier, deep dive analysis at your fingertips! :)

Pivoting

If you are an Excel geek, and I might qualify for one :) , you know what pivoting is all about. But for the purpose of this post, pivoting in Google Analytics will allow you to see additional metrics in the same view.

For example, say you are looking at your top landing page report. With secondary dimensions, you can now view the visitor type (new versus returning) as well.

Before pivoting in Google Analytics

This above report is for a news website, “/” is the homepage, and “/Politics” is the politics page. We see that the bounce rate for the “/Politics” page is much higher for new visitors than for the Returning Visitors. Time for action! Equipped with the new findings, you can review the “/Politics” page content and/or layout and assess how to further engage the new visitors. Keep in mind that when you are doing this type of analysis, keep statistical significance in mind; don’t waste time on something that is not statistically significant such as a seldom visited page.

With pivoting, the deep dive analysis is about to go into over drive. So while I am in the same GA report, it occurred to me that the client makes frequent updates to their homepage and maybe some browser incompatibilities have been introduced along the way. With a couple of clicks, I can get the insight I am looking for.

In the Secondary Dimensions drop-down, I selected “Browser”, then I selected the “Pivot” view and I choose “Operating System”. Here you go, all the cool analytics data you want right here in one table. We are now seeing:

  • Home page (our landing page in this example)
  • Viewed by browser type (IE, Firefox, Chrome, and more)
  • Viewed by Operating System (Windows, Mac, and more)
  • By Entrances and the respective Bounce Rate
  • Wow, a lot of numbers to view, but the report is much more insightful and there is so much context!

Pivoting in Google Analytics

What do I do next? Easy! Meet with the web design team, share the data, and hopefully help the team prioritize fixing browser incompatibility issues starting with Firefox on Mac, and then Safari on Windows. Obviously, if you are not happy with the 34.27% Bounce Rate of traffic on Internet Explorer, then you’d want to allocate time to improve it on this segment of traffic since Internet Explorer represents a significant percentage of the total.

So to summarize, the secondary dimensions and pivoting features in Google Analytics allow us to dig much deeper into the data, and all done on-the-fly. Give these features a try and let us know what you think.

Related Posts:

Technorati Tags: , , ,

May 05
2009

A very basic but extremely vital part to web design is understanding your users’ logical thought process and flow. You want to take this into account as early as possible in your design phase so that you create a very familiar flow to the user.  Putting a link or information in the next logical place makes the process intuitive and thus, more likely will lead the user to convert.   The more familiar the flow, the less they have to figure out where to go which betters their chances of getting from their starting point to your ending point.  The your end-point could be them submitting your form or buying your product.

While missing this boat can make you question “Why isn’t my page converting?”, getting it right can change your page from an occasional “hit-or-miss” to a conversion machine.   In some cases, simply changing your design to make a button or link more intuitively available might end-up increasing your conversion rate by 100′s of percentage points.  That small design update just grew your business astronomically.

There are two main aspects when it comes to understanding logical user-flow:

  1. User-interface and page yayout – actual elements of the page that the customer will be interacting with.
  2. Site structure and navigation – how your pages are organized.

In this post, I’d like to go through the first item – UI and page layout.

User-Interface and Page Layout

As I said, it’s important to remember the way your user will interact with your page while designing your page layout.  Take into account your users’ natural reading pattern – which in the U.S.A is left to right, then top to bottom. Thus, the top-left-most content will be the first thing they see then and they’ll be moving in an “F” pattern down your page.

Here’s an example of a login form that does poorly at following user-flow:

Poorly designed Control Center login form.

Can you spot the issue here? ANSWER: The “Login” link is in a bad place (you see it at the top right).

  1. A user will go to the “Customer ID” field and enter their ID.
  2. Moving left to right, they will move on to enter their password.
  3. Resetting now like a type-writer (I don’t know if people still remember those), the user will move to the “Language” field below.
  4. Ok, now were ready to login. When I move down and to the left…where the heck did the “login” button go?

The placement of the login button is not in the next logical place, and not only breaks up the user flow, but is also hard to find (aside from the fact that it is not a clear clickable button).

Now this may not seem like a big deal here cause there’s not much going on in this form. But what happens when the form is longer down the page and they have to scroll back up to search for this login/submit link? Should we assume the user knows to do this?

No. You’ll be surprised the percentage of people clicking such a link decreases or increases based on simple logical placement of this button.  Remember, a as surfers, we do not want to think.  We do not want to search.  It is as easy for us to close the page, and move on to the next site in a blink of an eye.

Worse – let’s say this is a “Checkout Now” button and your site makes $10k a month in revenue on this product. If the placement of this button even affects conversion even  by 10%, that’s enough to make a difference.  Larger sites that make significant online revenue a month would be greatly impacted, where even a small percentage of decrease would hurt (or a small increase could pay-off big time).

In the case that you are a smaller/medium size online business, online conversion may be your businesses only means of survival, so a small design detail like this could even make or break your business.

Technorati Tags: ,