3 ms·

The headline is a little misleading: they got rid of annual performance reviews in favour of more frequent performance discussions. I'm working for a startup b

The headline is a little misleading: they got rid of annual performance reviews in favour of more frequent performance discussions. I'm working for a startup building a performance plan/review SaaS app and we've always tried to encourage customers to use 3-4month reviews periods — but almost all are wedded to the idea of annual reviews. If you're stuck on an annual cycle, at least make sure the manager and employee have regular meetings so those goals and the current progress are frequently in the forefront of their minds. There's no point getting to December and then remembering what you and your team were supposed to be doing Don't set 2 A4 pages full of "goals", including unverifiable stuff like "maintaining a happy, positive atmosphere" like my managers at IBM did. Set 3-4 clear goals with defined metrics (hint: if there's not a metric to measure, it's not a goal). Some goals are also amenable to breaking down into milestones.

garenp 13y ago
The problem with setting very concrete goals like this, is that whenever you don't precisely meet them, they can be made to be a knock against you. As a result, it's then in your interest to set goals that are as easy to achieve as possible. I can't imagine that aiming for mediocrity serves the company or the employee well.
- hopeless 13y ago
  And setting vague goals which may, or may not, have been achieved is better for the employee or the company?! But you're right to an extent, and I skipped over it. There's a class of things (call them skills, behaviours or competencies) which aren't suited for evaluation through measurement: "teamwork", "leadership" etc. These are not goals and it's important to understand the difference. For competencies it's best to evaluate them on a simple scale (say 1-10, we don't even label the exact score) so the discussion naturally becomes a bit fuzzy and less concrete — because the concepts are fuzzy. Let's be honest, you want employees to improve in these competencies but there's really no effective way to evaluate them. Plans and reviews should be driven by the measurable stuff (goals) and the employee's development is driven by discussions not scoring of their skills/competencies/whatever-you-want-to-call-them
- hopeless 13y ago
  Also, the outcome when reviewing a goal doesn't have to be a binary achieved/failed. You could have something like failed/nearly-met/met/exceeded/wildly-exceeded
- rpwilcox 13y ago
  Yup. And after getting dinged once or twice in a review, metrics get gamed consciously or unconsciously. The mind tends to avoid pain. If management lays down some metric like "higher velocity by 20%" or "20% more bugs closed", don't be surprised when they get it and it turns out to not really be effective. (From imperceptible Story Point Inflation to 20% more technical debt, or bugs mysteriously skipping the bug tracking system...)

garenp 13y ago

The problem with setting very concrete goals like this, is that whenever you don't precisely meet them, they can be made to be a knock against you. As a result, it's then in your interest to set goals that are as easy to achieve as possible. I can't imagine that aiming for mediocrity serves the company or the employee well.

hopeless 13y ago
And setting vague goals which may, or may not, have been achieved is better for the employee or the company?! But you're right to an extent, and I skipped over it. There's a class of things (call them skills, behaviours or competencies) which aren't suited for evaluation through measurement: "teamwork", "leadership" etc. These are not goals and it's important to understand the difference. For competencies it's best to evaluate them on a simple scale (say 1-10, we don't even label the exact score) so the discussion naturally becomes a bit fuzzy and less concrete — because the concepts are fuzzy. Let's be honest, you want employees to improve in these competencies but there's really no effective way to evaluate them. Plans and reviews should be driven by the measurable stuff (goals) and the employee's development is driven by discussions not scoring of their skills/competencies/whatever-you-want-to-call-them
hopeless 13y ago
Also, the outcome when reviewing a goal doesn't have to be a binary achieved/failed. You could have something like failed/nearly-met/met/exceeded/wildly-exceeded
rpwilcox 13y ago
Yup. And after getting dinged once or twice in a review, metrics get gamed consciously or unconsciously. The mind tends to avoid pain. If management lays down some metric like "higher velocity by 20%" or "20% more bugs closed", don't be surprised when they get it and it turns out to not really be effective. (From imperceptible Story Point Inflation to 20% more technical debt, or bugs mysteriously skipping the bug tracking system...)