It's totally possible that there was a ton of bloat. It happens when organizations scale and they're looking for market share instead of profitability. But, as someone who has worked in (bio)tech companies that got purchased, first they cut 20-30% of the staff, but that doesn't mean all the work they were doing went away, and it means key people in every role have to do more, which works for a minute, then management thinks "we're right, it was all fat", until those key people burn out and stop producing. Then you end up hiring 10-15% of the staff back. Or, the org slowly dies.

> as someone who has worked in (bio)tech companies that got purchased, first they cut 20-30% of the staff, but that doesn't mean all the work they were doing went away, and it means key people in every role have to do more

Yeah, this wasn't a simple staff reduction. Twitter fired something like 80% (?) of their employees. By conventional wisdom, they should have gone down hard and never recovered. There are lots of people in this comment thread who are (dubiously) trying to make the case that they did.

The Twitter example, for better or worse, revealed that tech companies are not just a little bit redundantly staffed...they're employing tons of people who effecitvely just make work for each other.

> There are lots of people in this comment thread who are (dubiously) trying to make the case that they did. > The Twitter example, for better or worse, revealed that tech companies are not just a little bit redundantly staffed...they're employing tons of people who effecitvely just make work for each other.

I've been arguing since the purge that Twitter was in the "Wile E Coyote ran off the cliff and is still treading air" phase, but that argument is getting harder and harder to justify as Twitter somehow stays alive. How could firing all those people really do nothing to the operational success of the company? What were they all doing? Companies don't hire people where they have nothing for them to do. I look at my (BigTech) company and everyone is running around like crazy with 3-5X more tasks in their backlog than they can possibly do. We always need more people. How do you et rid of 80% and just carry on???

Yeah, that's why I was careful about my wording. Without excluding the possibility that many people were truly doing nothing, I see this pattern all the time as companies get bigger:

  * project is done to 85% effectiveness by <= 1 FTE

  * company is flush with cash, and can hire someone to backfill that last 15%

  * new person is not fully occupied backfilling that 15%, and therefore *finds new things to do*...which creates more projects that are done to 85% effectiveness

  * goto step 1

Also, of course, as companies mature, they inevitably start moving down the cost-effectiveness graph, and taking on projects with lower and lower expected value. There are all sorts of reasons for this, ranging from valid ("this effort might pay off 10x"), to pathological ("if my team doesn't grow, I don't get a promotion").

Anyway, shovel enough money into the boiler during the "hypergrowth" phase, and you can easily end up with lots of people who are all very busy, but just not productive.

I also thought there was a decent chance the entire product would just die irrecoverably, and that hasn't happened (although of course there might have been near misses we don't know about). I must admit I'm surprised how well it still seems to run, on the whole.

I think it's still an open question whether the business is damaged irrecoverably. I guess now they don't have the same reporting requirements, as a private company, but everything I've read suggests they're losing as much money as ever.

Although... you can certainly imagine a world where they both avoided infrastructure-death and managed to sustain the business (not pissing off advertisers, not pissing off power users, no pointless rebrands, etc).

Great. This is all completely besides the point anyway. My point is that there are more of them and I think anyone who doesn’t see that is nuts. I’m sorry I don’t have a Grafana dashboard to show you.

The point is, without a Grafana dashboard, you don't have anything but a motivated anecdote.

I never pretended to have anything else, and “motivated” is a weird word to use here. I have no skin in this game, and from what I can tell neither do you. If Twitter survives, great. I guess some of what you noted could be an interesting lesson, but I don’t think it’s going so hot so far and I’m not willing to call it yet.

You have a belief that the service is more buggy, and were arguing as such from anecdote. That's what I mean by "motivated".

I don't care at all, and am just reacting to the phenomenon of folks going "it's definitely worse now, and I know, because I used the product before". OK, great...I used the product before too, and don't perceive that to be true.

~90% of the comments here are "motivated anecdotes" and are quite valuable. Perhaps this is not the dis you think it is.

"motivated anecdotes" aren't valuable because they aren't accurate. It's political fuel for to shit on something you don't like.

With this line of thinking, I could say climate change isn't happening because of the temperature outside my house.

uhh visible product changes as proof that a company is more productive is a joke. A joke PMs love to tell at least lmao.

Making visible product changes is the easiest thing to do in software.