[QuercusMax]

If GRAD were actually designed to make layoffs easier, they would have put in a stack ranking feature. As it is, it puts more people in "safe" buckets than would be there in the old system, so I'm really struggling to see how this is layoff-related at all. They very clumsily instituted last-minute "support checkins" for people who were struggling, because they forgot to deal with the scenario as the entire process was rolled out exceptionally badly.

Laying off contract employees is extremely expected; that's a major reason why they hire these roles in the first place. I know that in the past when projects have been cancelled, the contractors have been let go almost instantaneously (like the next business day), but "real" employees have been kept on and helped to find new roles.

[Arainach]

This depends on who you talk to and what you consider "safe". I don't believe that GRAD has anything to do with layoffs, but I've absolutely talked with managers (including some who used to be managers at companies with stack ranking) who say that the wider lowest bucket adds pressure for managers to put at least some of their folks in it (before with the tiny lowest bucket it was more statistically justifiable if no one got that rating), which feels (to them and to me) like stack ranking.

[joshuamorton]

I think a lot of this comes down to explicitness. Just-barely-cme was always a thing, and people knew about it. Making it more explicit and forcing managers to justify it to reports and to other managers is actually good in a lot of ways.

The new larger bucket could be bad, but as far as I have seen getting it is a signal that you're at risk of being at risk, not that you're gone. And there's enough people who have survived PIPs and gotten NI without getting pipped or fired that I have some level of trust for management here.

[QuercusMax]

That's how I've always viewed it as well. Used to be a very large CME (consistently meets expectations) which included people from those who were struggling, but not enough to warrant official intervention in the form of a PIP (performance improvement program) or NI (needs improvement) rating, all the way up to those who are actually doing a solid job, but in an extremely average way. Kinda like a "gentleman's C", where you really should be getting D's, all the way up to maybe a C+. I think the idea is with a much narrower "pack your bags" type rating for those really in trouble, and a decent sized "you need a kick in the pants to get your act together" bucket, you'll be able to actually help more folks perform better.

I can see how making the "you're in danger of being fired" rating more explicit, combined with expected distributions (which we've been told are not enforced, but rather a sanity check, and only expected with orgs of a large enough size) of ratings, might look like you've designed a system to make it easier to lay people off, though. But this is assuming that it's all an elaborate smokescreen to make layoffs possible, when in reality if Google leadership decided to, they could absolutely lay people off. This new system has been in the works for the better part of a decade at Verily and X, it's not a reaction to the current economic system. I think honestly it's just that nobody knows how to do this stuff well, and people in HR are motivated to make changes to show that they're doing something and aren't just playing on their phones all day.