It’s Okay; We’re Confused, Too, Part 2

For many years, I’ve worked in IT for the headquarters of a major craft retailer in the US. The company has recently switched to a new Incident Tracking system. The system is used to log problems, escalate them to the right people, and, most controversial for us, track the time it takes to resolve a given issue. They make a big deal out of how this will allow us to be compliant to an industry standard known as ISIL.

Where the problem comes in is that the standards for what is called the “Service Level Agreement” or “SLA” are set by our notoriously broken help desk. As a result, they tell us that the more severe the problem is, the faster it needs to be resolved, which sounds good on paper… except…

They arbitrarily decide on a length of time that is acceptable for each SLA and punish broken SLAs. As a result of highly questionable resolution times, this results in most of IT failing to meet the set SLAs, which are sometimes as little as an hour for a major issue with no known cause.

Needless to say, there are a lot of unhappy people who are complaining a lot, and morale — which was already low for a number of other reasons — is falling again. Management decides that the real issue is that we just don’t understand the new software and how to use it. If we did, we would understand how to address these SLAs. So, they hire an ISIL expert consultant to come and give us a number of MANDATORY training classes.

These classes are spread over several days, and the trainer is actually quite good, but at the end of the session I attend — which is also attended by the VP who insisted on this software — he asks if there are any SPECIFIC issues he can help us understand.

Oh, boy.

Hands shoot up all over. Naturally, the questions are all on Service Level Agreements and the time limits on them.

The trainer nods agreeably and explains. 

Trainer: “Well, you see, an ‘INCIDENT’ is always the final category. Anything that you haven’t diagnosed should be entered as a ‘PROBLEM.’ The ‘Problem’ category is what all calls should initially come in as. SLAs only come into play as an ‘Incident,’ and the SLAs should be easy to keep track of.”

Please note that this is EXACTLY the opposite of what we have been told. There is silence in the room. People keep glancing at the VP whose baby this is and at the Help Desk Manager who devised all the training and SLAs for us. Finally, one of the bolder analysts raises her hand.

Analyst: “So, what you’re telling us is that SLAs should not be set for anything that is initially coming into the queue, and should only be set when we’ve determined what the issue is?”

Trainer: “Exactly!”

Analyst: “Because right now, we are given SLAs for every problem that gets opened.”

Trainer: *Looking a bit shocked* “Well, that’s just stupid. How can you possibly know how long it’s going to take to fix before you troubleshoot it?

Analyst: “EXACTLY!”

She shot a look over to the VP and the manager, who were looking a little uncomfortable.

The next day, we came in and saw an email that all further ISIL training classes were suspended, and that we were expected to use the Incident Tracking System the way we had initially been told. The VP then tried to spin it, “We didn’t need those classes; after all, we’re not an ISIL shop.”

Related:
It’s Okay; We’re Confused, Too