Every IT professional has a trove of war stories about challenging customers or end users. These “problem” users are not of a homogeneous sort—I can think of a half-dozen classifications off the top of my head. But today I have one particular breed in mind: the user possessing too much knowledge of an application. He might have been an implementer in a past life. He might even have been part of the project team that established the system. Sure, there is a sub-species of this user whose knowledge is dated—tied to an ages-ago release that has long since changed. But we are concerned here with the other sub-species, the user whose knowledge is current; he might even have an active user account on one of the pre-production environments. He is a most dangerous creature.
“Our old boss was like that,” one of my product managers told me. “We had to take her access away.”
For our implementation of Hyperion Planning, I have become “that user.”
The deployment project ended in November 2010, but this is my first year using the system to model my own departmental revenue and expenses. It has been hard, for all the reasons that budget-building and performance-monitoring are challenging activities. But old habits die hard—and I can’t help but generate lists of potential solutions. “Couldn’t we do X?” I ask. “Check out the nifty form I configured on DEV… When can I get it on PROD?”
I try to stop myself. I preface every email or call with a heart-felt admission that I know better. In the end I am my own worst enemy; I helped write the change management rules, to make sure that enhancements are treated with due rigor: formally defined and prioritized, tuned for extensibility and performance, integrated into communications and education plans. One-off solutions are dangerous—I know this and have preached it to hundreds of clients through the years.
But I can’t stop pushing.
Thank goodness my product manager holds the line. “This is great feedback,” he says. “I’ll include these for discussion at the next working group.” I can accept that.