User Experience

I wrote this post yesterday and lost it due to some random combination of problems with WordPress, Chrome and my laptop.  It’s ironic because the fact that I lost an hours work, and maybe key ideas that I won’t be able to remember didn’t get flagged to my companies service desk.  I didn’t ring the help desk, my laptop didn’t proactively report that Chrome stopped responding or log the fact that I had to hard power off my laptop.  Service desk analytics see only the very tip of the iceberg of user experience, but you’d think to listen to them that they see it all.

I’m fond of drawing this in terms of pie charts (but I will spare you).  The first chart would show that 99% of the time people are using their PCs and devices.  About 1% of the time things go wrong (apps are constantly crashing on my iPad for example) and we just carry on, we don’t contact anyone, our low expectations have been set.  No one tracks the direct productivity impact, the cost of distraction or the lost opportunity cost of this less than optimal experience.  It’s rare that anyone asks whether it can be improved (actually Apple, Microsoft and Google et al do collect ‘customer experience data’ but service desks don’t get to see it).

Now when we are motivated to action (perhaps 30% of the time) we still don’t contact the service desk, we ask the guy at the next desk, search on google, or ping a message to our personal IT guru.  Only when we are really ‘desperate’ do we contact the service desk, maybe 10% of the time.

In numbers then 99% of the time IT works, 1% of the time it doesn’t, 0.3% of the time we do something when it breaks, but only 0.1% of the time do we contact the help desk.  So ‘IT’ views 0.1% of my experience.

So I’ve been asking myself have we got things in balance,  I think not

we invest $millions in supporting the broken service (0.1% of the users experience), but $thousands in improving the usage experience (99% of the users experience)

One example of being out of balance is the service portal, which is a great step forward, but because it only supports 0.1% of the users experience it’s not a part of their working day, people don’t know there way to it, or around it.  I’ve been wondering whether we should be helping our users more, providing them with a companion application on the desktop that supports them in their working lives, as well as there non-working lives.

I’ve also been thinking a lot about the Pareto Principle, often known as the 80/20 rule, which postulates that 20% of the input yields 80% of the output.  In the context of support where there’s often a very long tail of issues and the cost of addressing many of them is high we might think of this as the 30/5 principle, ie if we could ‘fix’ 5% of the issues (the common, easy to diagnose and easy to fix ones) then we might address 30% of all issues.

Taken together I’ve been thinking about creating an application that enhances the complete usage experience of PCs (virtual and physical) that adds value that can’t be delivered by web based tools, that’s quick, easy, and fun to use.  It’s ambition is small, only 5% of the problem, but it’s impact could be big eliminating 30% of the work.

Steve and I had a great discussion about what this might mean in detail yesterday and Simon has produced a great mock-up in PowerPoint.  The best part of yesterday’s discussion concerned how would we create a client application that was self maintaining, trivial to configure for different customers, but had powerful extensibility.  We used some of the standard patterns of the web for this discussion, Podcasts, RSS, XML configuration files that could be rendered with the addition on CSS to create knowledge articles.  It was great fun.

The coolest part of the discussion concerned how useful and trivial it would be to help users diagnose and resolve the dreaded ‘is it the network?’ problem by providing a robust definition of how to test that LAN, WAN and Internet connectivity were working and that the device was connected to a network that actually had connectivity to the service the user wanted to access and that service was functioning.

It was great fun.