“It’s mid-year career discussion time at Microsoft®. I could rant about the HR tools we use, but that’s like complaining about prostate exams—too inflated a target. Instead, what gushes out at me at this time of year are BOGUS commitments.

You’ve heard of SMART goals (Specific, Measurable, Achievable, Results-based, and Time-specific). BOGUS commitments are Bloated, Outdated, Generic, Unrepresentative, and Self-centered.

What really kills me are BOGUS commitments posing as SMART ones. They sound specific and measurable. The results seem achievable in the specific time period provided. And yet, these so-called SMART commitments are totally BOGUS. As a manager, BOGUS commitments are particularly gut wrenching to evaluate—“Yes, you are ‘On Track’ for 11 of your 12 commitments. Unfortunately, your commitments make you look good while your team is failing, nine of the commitments have changed, and the one commitment that needs improvement is the one that matters the most, but it’s lost in the mess. AHHHH!!!!”

How do SMART commitments become BOGUS? Let’s break it down.

Eric Aside

To its credit, HR has heard all the “feedback” on the performance tools and is actively working on them. Unfortunately, the problems are systemic, so the fix won’t be quick. If you’re one of the Microsoft employees given a chance to provide constructive feedback on HR prototypes, please do—for the good of us all.

I get bloated with a foamy latte

The first trap for even the SMARTest of commitments is bloat. You’ve got a whole year’s worth of work. The commitments are supposed to be specific and measurable. They are supposed to align to your manager’s commitments. Pretty soon, your number of commitments is blowing past 10 and pushing toward 15.

Why is having 10+ commitments a bad thing? Let me count the ways:

1. It takes a long time to write them all. The goal is to get real work done, not to spend weeks writing and editing commitments.

2. It takes a long time for all of them to be reviewed regularly, or discussed at calibration, so they aren’t. So they are useless.

Eric Aside

Calibration is a process Microsoft uses as part of its differentiated pay system to calibrate expected and exceptional contributions and potential of a group of employees in comparable careers stages and roles. For example, the development managers for Office® meet regularly to separate all the Office developers within each career stage into the top 20% of high potential contributors, the middle 70%, and remaining 10%. Doing so aligns differentiated pay and our expectations of roles across the company. When used carelessly, calibration can emphasize local versus global optimization, but it’s the system we have.

3. They get so specific that the ones for the second half of the year are completely outdated, whereas the ones for the second quarter are only mildly outdated.

4. Since there’s no way to distinguish the critical commitments from the “that’s part of the job” commitments, you and your manager can’t give proper attention to serious problems.

5. Likewise, with countless commitments, it’s difficult for you and your manager to highlight your greatest achievements on your review or at calibration meetings.

What’s the solution? Have, at most, four commitments (five if you are a manager). Your first three commitments are for your critical project responsibilities this year—the ones your boss will either brag about or have to defend in calibration meetings. Your fourth commitment should focus on personal growth. (Be sure to include mentoring and teamwork skills in this one.) If you are a manager, your fifth commitment is to managing the health and growth of your team. Five commitments. End of story.”

So begins I. M. Wright’s “Hard Code” post and 8-minute-52-second podcast this month. (“Hard Code” is “An opinion column for developers. Brutally honest, no pulled punches.")

We hope that you’ll also check out I. M. Wright’s “Hard Code,” by Eric Brechner (Microsoft Press, 2008). Eric Brechner is Director of Development Excellence in Microsoft’s Engineering Excellence group.