The goal of getting IT right — making sure our technology works, and works for the benefit of the business — has never been tougher to accomplish.
Fortunately, one thing is still easy: screwing up.
What? You’ve forgotten how? Never fear! Just commit this list of a baker’s dozen tips to memory — or clip it out and keep it handy — and you’ll never be short of bad ideas again:
Assume. Assume you know what users need. Assume you know what managers expect. Assume you’ll catch up when the schedule slides. Assume no problems will show up in testing.
Expect. Expect support from management. Expect perfection from vendors. Expect clear specifications from users. Expect flawless execution.
Overpromise. Paint an improbably beautiful picture of glitch-free hardware, bug-free software and friction-free networks. Set an early delivery date.
Complicate. Avoid straightforward designs. Sneer at simple solutions.
Alienate. Alienate the users who can explain to you what your systems need to do for them. Alienate the business managers whose budgets you’ll spend. Alienate senior management, from whom all funding flows.
Experiment. Users make the best guinea pigs. Production systems make the best testbeds. Untried technology offers its greatest rewards in mission-critical applications.
Deny. Deny responsibility. Deny promises you made. Deny failure. Deny the limits of your staff, your systems and your own abilities. Deny what’s obvious to everyone else.
Change. Variety is the spice of life, and everyone loves something new. Change what users see on their screens. Change their passwords. Change what gets blocked by your filtering software. Change your help desk hours. Change your procedures.
Procrastinate. Delay that status meeting. Push back that code review. Put off those patches and upgrades and audits. Isn’t there always time to get things done at the last minute?
Withhold. Withhold information. Withhold cooperation. Withhold anything that would give users, business-side managers or executives a better sense of what to expect.
Underestimate. Make low guesses for cost. Predict a minimum of problems. Sketch out short development schedules. Ignore. Ignore problems, warning signs, complaints and objections. Why sweat that small stuff? After all, you’ve got IT work to screw up.