Yeah from what I've heard you ideally want to hit 70-80% of your OKRs, and people game it to make sure they fail at one or two so they don't get accused of being "too easy".
One of the most hilarious things I've ever seen was the head of Google Plus loudly sharing his "1.0 OKR" regarding social adoption at TGIF. It was about that time folks got suspicious and some long-termers found out Vic was lying about adoption rates.
Oof. Is that really why he got canned. I'd always assumed Gundotra just made too many enemies during the period where L&S empowered him to do whatever it takes to win in social, and when he didn't win, he had burned too many bridges to stay. Lying about adoption is something I didn't hear before.
I’ve seen this in many companies other than Google and shockingly even at startups.
If setting an OKR is meant to focus the team and maximize effort towards solving those problems, this approach is counterproductive because it completely fails to measure the effort exerted towards a goal.
You could have a 1.0 OKR and you could have 2 cases.
1. Set it too easy, didn’t have to do much to achieve it
2. Set a hard goal and produce a Herculean effort to achieve it
The latter case isn’t accounted for. It’s is glaringly obvious but the dull manager types don’t seem to want to acknowledge the difference or the fact that the latter behavior, if incentivized, leads to better and more predictable outcomes.
Instead, the effort is met with a blanket: “Too easy, didn’t set a hard enough goal”.
This incentivizes people to set easier goals that they can meet comfortably and slack of 20% of the time so that it doesn’t look like the goal they set was too easy.
I've never seen or even heard of anyone throwing their OKRs to avoid the appearance that they were too easy. In fact, you rarely hear about another team's grading of OKRs at all. Plenty of teams inside Google also set OKRs expecting/hoping to hit 1.0 so it wouldn't be at all surprising to see lots of 1's/near 1's on teams.
My team (under ads SRE umbrella) usually did aim for ~0.8 with a 1.0 stretch goal of some sort. It was a fairly reasonable calibration to make sure we were scoping and planning OKRs accurately.
If every OKR got 1.0 it meant we could comfortably take on more work next half, below 0.8 and we would plan to do a little less next half.
In theory it would have been fine to score OKRs above 1.0 for stretch goals for the same effect, but the software didn't work that way.
Well this was told to me by xooglers who were now working for other companies, so either they were the ones doing it and that's why they left, or they made it up to make Google sound worse. So I guess take it with a grain of salt?
12 years ago OKRs were huge. And yes, you graded them, managers and other teams viewed them, and you aimed for 0.7. Recently, it was a lot more lax. Many teams didn’t do them, drifted away from them, or didn’t bother grading them
Here here! We don't have to believe everything internet strangers say. The presupposition that an unvetted internet comment will somehow become "vetted" by the probing of _another_ internet stranger doesn't make any sense.
That culture came straight from Larry and Sergey. When I was there they would say at TGIF explicitly that if you score 1.0 they'd get suspicious because maybe you were setting goals that were too easy. Guess what, nobody ever got 1.0