These are not valid metrics of the health of a website. These are numbers that executives like to hang their hats on, and they are shortcuts to thinking and in-depth analysis. I love metrics, no doubt, but it’s time to start telling stories of user-interaction. Hard numbers lack context; and therefore often send poorly informed decision makers down the wrong path. Let’s go over a few of my favorites.

Absolute Unique Visitors
This number is by far the most used metric of site success but it needs to be framed in the context of goals & objectives. Look at the weekly trend below. It’s a capture of our traffic pattern on supportdetails.com. Support Details is a one-page site that captures browser specs and sends them to a targeted contact via email.

Simply showing this graphic and looking at an absolute number doesn’t tell the story of what the users are doing on this site. When this graphic is supported by referral sites, repeat visitor trends and normalized for spikes then the numbers begin to talk.

The numbers are telling me the spikes in traffic have helped to create product awareness, overtime this increased awareness has led to increase usage of Support Details on a regular basis. By utilizing the visualization tools that Google Analytics has to offer the trend becomes pretty apparent.

Pageviews
More pages clicked is counter-intuitive to the rational of a good site
information architecture. So why is it valued as a positive metric? If someone is clicking on more pages it can either mean they aren’t able to find content, or it could mean they are deeply interested in the content. I usually suspect it’s the former.

Bounce Rate
Measured by a visitor coming to one-page of the site, then leaving. This isn’t necessarily an indicator of a problem. It could be a measure of excellent content. If a visitor is searching for information and they find it without clicking around, then wouldn’t this be considered success? When this metric is used in the context of goals, then it becomes important. As a standalone number, it’s a useless metric.

Time on Site
What does this really mean? Does increase time mean the site is difficult to navigate. Is it hard for me to find what I’m looking for, or am I engaged in reading the content. In-depth analysis would compare the objectives of the site, with the type of content, then the average stay on a particular page.

I could continue offering alternative views for just about any of the major metrics, but my point is it’s time to look at the bigger picture. We need to be asking questions rather then looking for shortcuts that make for pretty charts and reports. Dig deep, give real analysis into what’s perceived in reading the numbers; that is were the art & skill of analytics pays off.

I’d like to hear more stories of useless analytics that you’ve encountered. Please share the info.

In the Beginning

When I first started using Mac OS X in college I fell in love with the Safari web browser. It was light weight, sexy, and fast. Unfortunately, the more I ramped into complex web development the more I realized Safari’s severe developer limitations. Back then developers had to enter terminal commands to even gain access to Safari’s developer menu. And, once turned on, the developer menu lacked the oomph of Firefox’s flourishing Firebug and Web Developer Toolbar extensions.

Since that day I’ve been using Firefox heavily for development and only occasionally jumping into Safari to test site feel and functionality.

Then, about a year ago, Safari started making real steps to support web developers. The release of Safari 4 earlier this year capitalized a nice upswing in Safari developer improvements. But, unfortunately, Safari still lacks some of the most basic development capability.

In my opinion, if Safari doesn’t address the following three issues, it will never become a viable competitor to Firefox for developers.

Source View Made for the 80’s

Any time your browser’s source code view renders like Internet Explorer’s you really need to sit down think things over. Is this acceptable, am I being an asshole? The answer, obviously, is yes.

Let Me View Background Images!

The web is no longer a mass of tables and img tags. Thanks to CSS we’ve started abstracting background effects from real content. Knowing this, there is absolutely no reason why Safari shouldn’t have a view background image option. Safari, do you really expect me to right click, inspect element, and then track down the CSS that links to that background image? Give me a break.

Target HTML Elements Faster

The best thing about the Web Developer Toolbar extension for Firefox is that you can hit cmd-shift-y and instantly view HTML elements. This is perfect if you’re trying to track down a styling issue or remember an ID name. It’s fast, easy, and accurate. Safari’s right click > inspect element isn’t even in the same ballpark as this. And, as hard as I’ve tried, there’s no way to reassign the Inspect Element shortcut in OS X.

Conclusion

Safari, you’re a great consumer browser. And yes, WebKit is a phenomenal web rendering engine. Hell, even Google loves you. Please, just fix these small problems and I’ll switch to you full-time, I promise! Until then, I’m going to continue buying FireFox t-shirts.

Recently I’ve been intrigued by the personal statistic service over at Daytum.com. In brief summary Daytum is a service that allows users to quickly generate graphs about anything they’re interested in. It then publishes them in a beautiful interface for others to see.

This new service got me thinking about how awesome it would be to have a detailed statistical overview of one’s life. I’m not talking bare bones stats, but rather in depth detailed statistics. I.E. cups of coffee consumed daily, total number video game hours played, the longest period of time spent awake, the number miles driven, the amount of jokes told, the amount of jokes flopped, etc.

Well, interestingly enough it was pointed out to me that the head graphic designer behind Daytum actually has compiled some yearly statistics of his life. You can check them out over at Nicholas Felton’s site.

Looking at Nicholas’ statistics made me start to wonder, how would you track these sorts of things? I imagine some of it would have to be guess work, maybe sitting down once every two weeks and going over your four or five big categories. “What was my average amount of sleep this week? How many girl’s numbers did I ask for, how many did I get? How much time did I spend reading?” Then, over the course of the year you could keep these life stats tracked in a service like Daytum. Or you could tweet them to a private twitter feed. Then at the end of the year you could run all this information through an excel sheet and see exactly where you stood. Granted it would be a definite time commitment, but I think it might be worth it to see an overview of your life in statistical format.

I’d be curious to know if anyone else has other good ideas for how to document these sorts of things, perhaps a good iPhone app? Leave your ideas in the comments or hit us up on Twitter @imulus.

Update: Edited spelling mistakes of Feltron vs. Felton, oops.

A few weeks ago I decided to set up a new service on my personal website to track all the social services that I’m a part of. Basically I wanted a central place where people could go to find out what I was up to, what pictures I was posting, and where I’d been. Now, I know there are services like FriendFeed and Socialthing! that do this on a larger scale but I don’t foresee many of my family members signing up for those just to see what I’m up to.

Enter Sweetcron, a free Wordpress like PHP and MySQL content system specifically built to handle lifestreaming. I heard about Sweetcron from my buddy Ryan and decided to give it a try for my new site. To my excitement it was exactly what I was looking for. Below I’ll cover the installation, interface, and end result of my Sweetcron experience.

Installation

Sweetcron is the easiest content system I’ve set up in quite some time. You download the compressed files and upload them to your server, create a MySQL database, and edit a few lines of a configuration file with the info. Once this is done the script has a built in installer that runs to finish the installation. By the time you’ve gone through this process you’ll have a username and password through which you can login to your Sweetcron administration section.

Interface and Use

Once you’ve logged into the back-end of Sweetcron you’ll see a few pretty basic options. Write: for posting quick notes, items: a list of your imported (published or unpublished) feed items, feeds: a list of the feeds you’re importing, and options: which contains basic account settings. Setting up your feeds to be imported into your lifestream is a piece of cake. Grab RSS from your favorite social networks and add them into your feed list¹.

Adding a Feed

Feed List

Imported Items

Sweetcron will automatically format and grab the necessary items related to the feeds. From there you can select from a few different themes of how you want your information to be displayed. If you’re ambitious you can set up your own theme by using the Sweetcron documentation.

Pulling Results via Cron

As the name suggests Sweetcron is reliant on pulling information from your RSS feeds at a regular interval in order to update your database. To do this Sweetcron works in one of two ways. First, you can use a pseudo method that will have Sweetcron update itself if the last update is older than 30 minutes. This method waits for someone to hit the site and then has that visit trigger the fetch. This may work for people without cron hosting access but I found the normal cron method to work better. To set this up I logged into my hosting administration and had the following cron job run every fifteen minutes:

/usr/bin/curl -s http://your-url.com/cron/link Note that sweetcron uses curl and this program may be located at different locations on different hosting providers, for instance:

curl -s http://your-url.com/cron/link

End Result

The end is a great looking lifestream that stores all of your information into a MySQL database that you host. The benefits are awesome. First, you own look and feel of the information output. Second, you don’t have to rely on a 3rd party service. And last, the whole setup is opensource, free, and easy to set up.

1. If you have issues with a particular feed’s RSS being supported I’d suggest routing it through Feedburner first and then importing it into Sweetcron. I had to do this with Brightkite.

Here at Imulus we all use glass top studio desks. There are a few main reasons for this: sex appeal, weight, and office personality. However, there are also a few hidden benefits.

From time to time when brainstorming and talking out ideas we’ll simply pull out a whiteboard marker and start drawing right on the desk. Rather than getting distracted moving into the conference room and drawing on the whiteboard (which usually involves erasing it first) we’ll just get our ideas out right away. Plus then you get the added benefit of being right next to the computer and you don’t have to retrace your ideas onto a notebook.

Sure, this could be done using a sketchbook or a portable whiteboard, but we find it a nice add-on to the standard “desk setup” of today’s office.