Skip navigation

Geek Speak

4 Posts authored by: davemhenry

In 1999 and 2000 I worked as Tier 3 UNIX Support for a colocation company. One of our largest customers had recently moved in so much new equipment, there wasn’t enough power in one rack. In what was to be a temporary solution, a power cord was left running across the aisle about two feet off the floor.

 

One day as I was going into the data center with one of our Tier 2 folks, the inevitable happened – his foot caught the cord as he stepped over it, pulling it free and crashing the customer’s server. He scrambled to plug it back in as quickly as possible, but I stopped him and instead had us find a longer cord which we then ran underneath the raised floor.

 

The outage triggered an automatic trouble ticket, which was assigned to me. My manager advised me to list the root cause as “momentary loss of power”. This in turn triggered a series of daily 8:30AM “post mortem” meetings as the customer – quite reasonably – wanted to know why it had taken so long to get the server back on line. I was instructed to not say anything that could make it appear that the outage was our fault in any way.

 

After six weeks, I couldn't take it anymore and, with my manager waving me off, I said, “I tripped over your power cord.” I then proceeded to tell the customer what had happened, making it sound like I had been alone in the data center.

 

The customer’s first response was, “Oh. You should have just said so in the first place.”

 

With the explanation that we had set things up so no one could trip over the cord again, the customer was satisfied with the resolution.

 

This incident led me to determine the three things any customer wants to know when there’s an incident:

  1. What happened.
  2. What you did to fix it.
  3. What you’re doing to make sure it doesn't happen again.

 

If you can provide (and follow through on) these three things, you’ll have satisfied customers nearly every time.

 

What are the key things you do to ensure your customers are happy with the resolution of their problems?

Any Help Desk is going to have to handle repeat requests (hopefully not always for the same customer – although repeat requests from customers is definitely a metric worth tracking…). There are a few things you can do to help avoid having “repetitive” become “mind-numbing boredom”, while improving the level of service provided to your customers.


The first, and most obvious, thing to do is automate. Many common requests received by the Help Desk can be automated through scripts. Requests like password resets, creating new accounts, permission changes, and provisioning new resources can all be automated.


Who will write these scripts? I’ve always found that this is a great job for the administrators who wish they were spending less time doing break-fix. Not only will they welcome the chance to keep their skills sharp by writing some code, but they’re also most likely the folks with the most knowledge of what needs to be accomplished to meet the particular request.


The second thing to do is document. Not all repeat tasks can be easily automated. Creating “How To” documents or “run books” for your Help Desk staff can make their jobs easier and keep your customers happier. The keys to successful documentation are:

  • Keep it centralized
  • Keep it up to date
  • Make it easily and quickly searchable


A lot of organizations find that setting up an internal Wiki serves this purpose well.


Handling repetitive Help Desk tasks well, whether through automation, documentation, or training, serves both IT and its customers. Help Desk feels more valued if they can help a customer quickly without having to transfer them to someone else. Customers feel better taken care of if the first person they speak to can assist them immediately. Lastly, the fewer repeat tasks that get pushed to other staff, the more time they have to focus on improving the overall environment.


How does your organization handle repetitive requests? What tasks do you see as good candidates for automation?

At some point IT will be asked to make a business case for some expense related to the Help Desk. It could be to justify hiring new staff, laptop upgrades, training, or to avoid the axe in a time of tightening budgets. When that time comes, for whatever reason, you’ll want to be able to show the return on investment (ROI) of the Help Desk expenses.


At first, this seems like it should be easy enough to show – simply calculate the cost of the Help Desk, create some metrics to calculate the value of what the Help Desk provides to its customers (customers, not users), and then demonstrate that the second value is greater than the first.


In practice, this is often extraordinarily difficult for two reasons. The first is that it’s actually hard to get people to agree on the value of the Help Desk metrics that are easy to measure. Some examples:

  • Number of tickets/cases handled: Some will argue that a high number proves the value of Help Desk, while others maintain that it shows the environment is too fault-prone or difficult to use.
  • The cost of the customer-hours of productivity that Help Desk saves: It’s difficult to get people to agree on the monetary value of a customer-hour or on how much time a particular Help Desk action saved.
  • Average time to ticket/problem resolution: A low number here is an obvious sign of a good Help Desk, but does a closed ticket mean the problem is actually resolved?


The second reason is that the things you actually want to measure turn out to be really difficult to put a number to. What you really want to know is:

  • Are our customers happy with the service they’ve received?
  • Are our customers more productive because of us? If yes, how much more productive?
  • Was the issue actually resolved satisfactorily, or did the customer simply work around it?


What metrics are you tracking for your Help Desk? What do you wish you could track?

Listen to IT administrators when they vent about their least-favorite aspects of the job and you’ll hear common themes: the long hours, time spent on-call, the thankless nature of the job, and the “this would be a great job if it weren’t for our users…” Dig deeper though, and just about every IT admin, if offered the chance to change one thing about their job, would change the same thing: spend less time on break-fix and more time implementing new projects.

 

Viewed in that light, the Help Desk can seem like part of the problem. It’s a never-ending source of trouble tickets that keep the admins too busy to get to the fun parts of the job.

 

I invite admins to view your Help Desk differently: it’s a source of invaluable feedback on how your department is doing. For example:

  • Have you set up alerts to proactively monitor your infrastructure? Problem reports from Help Desk will quickly let you know what you left off your list.
  • Did you roll out an upgrade or patch recently? A lack of any update-related problem reports over the next couple days lets you know you did it well.
  • Has there been an increase in complaints about network performance? Perhaps that network Quality of Service tool you’re using isn’t configured correctly to meet current need.
  • Has there been a noticeable decrease in reports of out-of-space errors? You must be doing an excellent job of capacity management and planning.


If you can view your Help Desk as a feedback tool, it will help you to serve your customers better. (I encourage you to consider them to be “customers” and not “users” – it will elevate the quality of personal service that you offer.) Better service to your customers leads to less time spent on break-fix, freeing you up for the fun parts of the job.


Administrators, how do you view your Help Desk? What’s the most valuable piece of feedback you’ve received through your Help Desk?

Filter Blog

By date: By tag: