In my last post I wrote about some of the issues for IT to consider in developing a successful telework or telecommuting program. One issue, in particular, deserves a little more attention: Instrumentation of your teleworking infrastructure.
Instrumentation adds monitoring mechanisms to IT services and is intended to achieve three main goals …
- * monitor and track performance
- * detect problems and errors
- * monitor and verify security
The first goal, monitoring and tracking performance, is critical because how well your services operate (such as throughput, trends, and peaks) tells you how the services are being used and whether they are meeting user demand.
The next goal, detecting problems and errors, is crucial in ensuring that remote workers doing their jobs don't waste time dealing with slow or buggy services.
Out of these first two goals comes the ability to establish benchmarks and do capacity planning. This is crucial because if your teleworking program is successful then it's going to expand and that will put more load on organizational resources. You will need this data to plan how you're going to keep your users happy.
The last goal, monitoring and verifying security covers everything from keeping the bad guys and their exploits out, through preventing the intrusion of malware, to ensuring that your teleworkers are neither unintentionally nor intentionally doing things to compromise your security.
So, you, the guys in IT, have got your infrastructure instrumentation all sorted out; you're tracking performance, correcting service problems, fine tuning and scaling resources … but there's one more aspect of instrumentation that is often ignored: What the users actually think, how they perceive the services, and whether they feel that the services help them do their jobs.
Sure, you know that Bob from pre-sales has messed up logging-in twice this week; that he didn't attempt to browse any blacklisted websites using his corporate laptop; that he's transferred X megabytes of email; and that he made Y requests to the corporate intranet web servers. But what does Bob actually think?
Does Bob feel that the IT services he uses get in the way of doing business? Is Bob frustrated with the content management system? If Bob blogs for the company does he feel the blogging system works well for him? If he doesn't, why not?*
This kind of instrumentation, surveying users, should be a planned, routine program, not something you do when you remember to.
You should sample your user base quarterly and poll not just random users but also the users who have had issues that you think you've addressed to find out if they are now happy.
And don't forget the in-person surveys. Sure, you can use online surveys via email or as pop-ups on your websites but don't forget to call users and ask for their input. When you can, gather workgroups together, give 'em some pizza, and find out what kind of job you're really doing.
Building effective IT systems that are properly instrumented isn't something that stops when you've done the technical parts … it has to include understanding the human dimension because that's what will determine whether what you deliver is merely good or out-and-out excellent.
So, have you talked to your teleworkers today?
* If he doesn't think it works well then he's probably using Drupal but that's a topic for another posting.