Wednesday, September 12, 2012

Google Analytics interface - designed by a team of sadists and software developers FOR smug software developers

Here's something I really hate - software developers designing user interfaces. And it happens ALL THE TIME. I really wonder how Jakob Nielsen makes a living selling all his usability seminars because I've yet to work for a company that does usability testing or gives a rat's ass about user-friendly interfaces. I think that Amazon probably has a usability team since the Amazon interface is very intuitive. But Amazon is the rare exception to the rule. 

The rule is that companies let the software developers create the user interface, as a by-product of creating the code. It's strictly an afterthought. And then they hire technical writers like me to explain their stupid user-hostile interfaces instead of spending money on designing an intuitive interface that doesn't require a manual to understand.

And Google, which has no excuse, clearly followed that rule. Because the new Google Analytics interface is obviously designed by a team of software developers and sadists. The sadists are responsible for telling you how cool all the features like Events and Goals are, and the software developers created the user interface for using these features that require you to be a software developer yourself to use them.

It's NOT USABLE. Unless you are a software developer. Don't believe me? Here's the Goal setup interface.





All I want to do is track when a link in one of my blog's HTML/Javascript widgets is clicked. You'd think they would already make that statistic available but no. You have to set up some torturous Event/Goal mechanism. The only reason I had ANY idea what "category" "action"  and "label" were is because I found some random guy on the Internet talking about it. It was here that I learned that I had to insert a piece of code into my A tag in order to create an "event." Here's the example the guy gave:

a href="http://www.webanalyticsblueprint.com/" onclick="_gaq.push(['_trackEvent', 'Ads', 'Sidebar', 'WAB']);"

Nowhere in the Google Analytics help section did I come across a heads-up that you have to use this  "_trackEvent" code. I mean, if nothing tells you that you will need to use code, you won't even be looking for it. 

I was able to glean some information because random Internet guy used an example. But his example doesn't include a value and his explanation of a value is 
"this element helps you specify a value for you event that can be used when you setup a goal for your event."
That's what is known as a tautology. It tells you that a value is a value. That's no help at all. But of course he's a software developer - he doesn't need to express himself clearly in English.

And to add insult to injury if you do an online search to try to find people to commiserate with about the fiendishly confusing Analytics interface, instead you get this:
Google Analytics already has one of the most usable interfaces in the Web Analytics world. 
 If Google Analytics "has one of the most usable interfaces in the Web Analytics world" what that tells you is that the entire Web Analytics world is composed of software developers and sadists.

People who look at web analytics are not software developers. They are people who want to make money from their web site. A web site that is not about software development. If they WERE developers they could create their own analytic tools. 

Now we've already seen that you have to have some level of comfort with code, but here's the best part of all: you have to learn something called "regular expressions." 

As another random Internet person says: Regular Expressions – Don’t Use Google Analytics Without Them.

Would you like to see an example of a regular expression? Here you go:


$string1 =~ m/(H..).(o..)/)

Now as it happens, I have some minor experience with regular expressions due to some database text verification work. I've also had experience with Javascript and SQL and PHP. I actually created a PHP page to track out clicks by sending the out click information to my SQL database on my ISP - but it took too long to process database insert so I tried to use Google Analytics instead. I almost ripped my own head off.

I can't imagine what someone with no code experience at all makes of the Google Analytics interface.

So anybody out there who is looking for someone else to commiserate with over the sadistic Google Analytics interface - you are not alone.