Hacker News Reader

Comment by SilverBirch | Hacker News Reader

More people == More politics. This varies by person, some people are highly political - you will not get any work out of them, except possibly in service of their political goal. It scales super linearly though, so at some point as you add more people you're actually reducing productivity, because those people are either going to spend all their time navigating the politics, which is bad, or start their own political agenda which is worse. You know what's bad? Hiring 1 unproductive person. What's wores? Hiring 1 unproductive person who persuades you to hire a whole team to serve their bullshit.

I've got good news though. How do you fix this? Simple, don't overhire in the first place! Womp Womp!

View on HN

> More people == More politics.

Not quite that pessimistic. I don't remember who but someone summarised it as "if you are willing to try many ideas at the same time, you grow innovators. If you are willing to try one idea at a time, you grow politicians."

Basically if you want to force the entire organisation to dance to the exact same beat, you'll get politicians competing for which beat to dance to.

If you accept some disparity and divergence in goals, and have decentralised systems and processes in place to loosely realign efforts, you can have many people and not so much politics.

(Is that inefficient? Maybe, but no worse than the alternative. Besides, perfect alignment of effort implies that someone knows what the future will look like -- never mind that different people have different opinions on this. If looking for a small trinket on a parking lot, would you rather illuminate with floodlight or laser pointer?)

Office politics is one somewhat cynical perspective (even if it is accurate as a hierarchy grows taller).

The other simple fact is that a larger team or a larger org requires more communication, more co-ordination, and more management.

Beyond a point, adding parallelism doesn’t improve performance, it kills it.

> Office politics is one somewhat cynical perspective

had director colleague say this in same words.

it cynical view, so we pretend it not happening and not rock boat.

well here we are now with layoffs. entire teams who produce no economic value. problem is too many people hired for empire building purposes.

it is easier to ignore real issues causing stagnating product. instead, bring in professional manager class. product owners, program managers, software engineering managers, senior mansgers. layers and layers of competing political incentive. building complexity for promotions and to increase head count.

Game they play is just say right platitudes like “obsessing with customers”, “all about customer journey”, “reduce friction for customer” and humble brag “my org have 120 people report”.

You're not wrong, but I think you're describing a separate issue too, which I also agree with. It's not politics, it's growing too big, too fast, and inventing jobs that were never needed before to meet some expectation of growth.

And, after all the layoffs this past year, it turns out that all the hyper-growth hires post-funding are just pawns to bolster a VC's investment.

Office politics in their pure form are rough, borderline toxic, like having someone threaten your promotion or your pay rise if you don't meet their sudden change of expectations.

There are certainly individuals with a problematic approach to team politics at Google (I have some stories). However I think this comment may put too much weight on the individuals without counting the overhead of ordinary, good faith conmunication. The more people you have working on something and trying to provide good ideas, identify and avert possible problems, and derive the right priorities the more stakeholders you have. The more stakeholders, the more process and communication overhead you will end up with. So even if you could screen out all the "political" and "unproductive" people you'd still be moving slowly in an organization with >100k employees.

Part of the problem is that actually... you don't want more people weighing in on the same priorities. If you hire someone either they should do something new, in which case they shouldn't really be weighing in on something existing as much as possible (certainly not have decision making power) or they should accelerate something existing, in which case they aren't a new stakeholder. If a team depends on your team that's 1 stakeholder. if that's 50 people, fine, but it's still 1 stakeholder. If 50 of them turn up to your "meeting with stakeholders" ask them which 1 is going to represent them. If 1 of them can't represent the other 50, ask them to leave and come back when they have a functioning structure, or fire 49 of htem, whichever is easiest.

In this scenario the team of 50 still needs to sort out what the one person is going to ask for at the stakeholder meeting and that probably requires additional communication within that team of 50, so even though it's contained within that team it contributed to the overall glaciosity of the company when it comes to getting things done.

at the bottom of the article they mentioned they hired 30k people to bring them up to nearly 190k last year, even if they perfectly fit in that's a lot of onboarding

That's actually low by Google standards, onboarding 16% of the company. There's been reports elsewhere that the average tenure at Google is 2.5 years, meaning a typical year sees onboarding 40% of the company.

40 % assuming a steady state of headcount. If they are growing, onboarding the growth happens in addition to that -- so 60 %.

Reply on news.ycombinator.com