[Southland]

> Our early analysis of performance data suggests that engineers who either joined Meta in-person and then transferred to remote or remained in-person performed better on average than people who joined remotely. This analysis also shows that engineers earlier in their career perform better on average when they work in-person with teammates at least three days a week. This requires further study, but our hypothesis is that it is still easier to build trust in person and that those relationships help us work more effectively.

[ASinclair]

> This requires further study, but our hypothesis is that it is still easier to build trust in person

I chuckled a bit. This is coming from a Social Media Company whose product is about connecting online. What does it say about the trust between users of its platforms?

[jayd16]

Well I mean, the simple explanation is online is still far better than nothing. I'm not sure anyone is expecting you to read someone's wall if you could just talk to them every day.

[enumjorge]

But given the choice, why would a company invest in VR headsets as a way to collaborate remotely if the company selling them thinks in-person is better?

[el_cujo]

I don't think those ideas necessarily contradict. If you're using VR to collaborate with someone on the other side of the country (or world) but you only need to collaborate intermittently, then remote collaboration just has to be "good enough" to offset the cost of travel even if in person would be ideal. Meanwhile, for their own employees who may need to collaborate daily, from their point of view they might think in that case it makes more sense to enforce in-person. I doubt Meta's stated goal (at this point) with their VR stuff is that you should never ever leave your house because VR is just as good, seems like it's more "this thing enables remote to be better".

[kapp_in_life]

It can be that in-person is best, but isn't always possible. And then it can also be the case that VR is better than connecting over Web/Mobile, while also being far cheaper than a plane ticket.

[jayd16]

Adding to what others have said, the headsets are actually MR devices. You can collaborate in person with shared virtual objects. I will admit that the software isn't there yet, but the vision is you could comfortably replace models and mock ups with richer virtual media.

[zeroonetwothree]

Originally Facebook was all about only connecting with people you already know. So it’s not as ironic as you are suggesting.

[gedy]

Yeah "it is still easier to build trust in person, so they can work together on collaborating via floating Wiimote heads in VR"

[Pyrodogg]

About as good as musk Tweeting that phone calls are better than tweets for actually connecting with some.

[myth_drannon]

Juniors and interns really strugle in a remote environment. Since Meta was hiring mostly new grads or junior leetcoders it makes sense that remote work culture at Meta hit some bumps.

[parzivalm]

> Juniors and interns really strugle in a remote environment.

Blanket statements like this are trying to make a multifaceted problem into a single problem. I don't understand why we are always so quick to say underperforming remote junior engineers are underperforming because of remote work. Maybe the problem was Meta went half in on supporting remote as an onboarding ramp and teams did not put the work in to make sure Juniors thrived in a remote environment.

Speaking from experience, If you are truly trying to be a remote company, you adjust as needed to support new employees, junior or not.

[saiya-jin]

I think you both are correct. Take normal graduate, split spacetime continuum into 2 parallel universes, and let 1 join same company remotely and onsite. Unless given company/team is very toxic, graduate who has gazillion questions every day and gets stuck even more will get them answered more easily if he just comes in person to seniors, rather than chasing calls/chats every time he is stuck or uncertain how to decide.

Quick brainstormings, pair programming. I mean its not up to discussion, real world on site is more efficient. Now from perspective of senior always bothered by those pesky juniors with their stupid questions, loud environment etc. its a different story. Company can set itself up to be much more open to fully remote, but its a conscious and continuous process that needs to be accepted by all seniors, people tend to revert to previous way of working.

Personally, I can see this also for seniors who are also onsite. In truly global teams, you are constantly chasing people and teams to do their work, approve processes, deliver stuff etc. Compared to person X who will respond to your email/chat in 4 hours it takes 1 minute including walking to get feedback, understand problem, manage expectations, push things further etc.

[I_AM_A_SMURF]

It really depends on a lot of factors. For somebody who's shy / socially awkward / introvert (lots of those in tech) it's _a lot_ easier asking questions in chat than in person, especially a shared chat with not too many people, it's also way faster than walking across the hallway to catch the person that you need.

In my experience coaching many young engineers in person and remote, the problem is not so much that it's easier for them to ask questions, is that they don't ask questions, like, at all. They are too worried to sound/appear dumb, even when all the senior/staff engineers are super nice. And in a way I understand that, they are unproven, they have to show their worth to everyone else.

My theory is that the idea that early career people don't do well remote is that it's easier for the senior people to forget about them and not check up on them regularly to make sure they're progressing. If a person is sitting close to you and not making progress it's a lot harder to miss. Also the casual "hey everything OK with your problem?" is a lot easier in person than online. That being said, with some adjustments it really doesn't take too much effort on the Tl/Manager side to make sure junior folks don't get stuck.

I've lead a few teams remotely now and the most successful ones were the ones where I started checking up on the most junior ones often, as often as every day, just asking low-key questions like "hey do you have any questions for me? are you stuck anywhere?" for a bit, just to make sure they were feeling OK asking me when they had a problem, and re-routing them to the right person as needed. After a while they feel safe and start doing it on their own, but it might take a while depending on the person.

As an experienced person, I can chat with anybody at the company within minutes, it's so much more productive than chasing somebody in the office (which could be easily 10~20min walk away).

In person is way better for socializing, creating team bonds, etc. but you don't need that every day, a yearly/quarterly team/company offsite is sufficient.

This is my perspective working at companies that are big enough that they would be global anyway. For a smaller company I can see the argument of having everyone in the same office, but even then your giving up a lot to make up for that (commuting, walking around finding the right person) and I don't think that scales well beyond the 1000 people mark.

[papito]

How many people out there randomly zoom with their former coworkers that they never met in real life, after leaving their job?

Your 20s are a critical time for your career to make connections and friendships - in person. You need to be around people, making friends over beers and community lunches, and learning social skills in a professional environment.

Are we that spoiled that the excuse is "I don't like public transportation"?

Do companies take a productivity hit when people are in the office? Probably. And many openly acknowledge that, but asking people to come to the office two times a week is not some sort of wage slavery.

[lopkeny12ko]

> Your 20s are a critical time for your career to make connections and friendships - in person. You need to be around people, making friends over beers and community lunches, and learning social skills in a professional environment.

Okay, sure. But why does this need to happen in the workplace?

And more importantly, why are you advocating for employers to unilaterally declare that this needs to happen in the workplace?

[MikeTheRocker]

Connection doesn't need to happen in the workplace, but the reality is that most people spend a majority of their weekday waking hours working so would be great if building connections there were feasible.

[parzivalm]

> Your 20s are a critical time for your career to make connections and friendships - in person. You need to be around people, making friends over beers and community lunches, and learning social skills in a professional environment.

No one is saying you don't need to be around people, but a job can be just a job. I don't need to make a ton of friends at work to do my job, and the idea that that should be a requirement/expectation needs to stop. It enables the chance for too many unhealthy boundaries to be created young workers that don't know any better. If you start in your early 20's and are made to see everyone as friends, family, etc. Then when someone tries to push you into a 12 hour day, it doesn't seem that bad. You can forge working/professional relationships virtually just as well, if you put the effort in.

Now sure it doesn't work for everyone, but again my argument isn't that one is worse than the other, it is just that we make too many blanket statements. What works for you might not work for me, and these organizations trying to force one or the other is harmful in the end. You can have co-existence, especially in a company the size of Meta, you can have a plenty successful office presence and remote presence.

> Are we that spoiled that the excuse is "I don't like public transportation"?

I don't think anyone said that in this thread... That being said, it shouldn't be a surprising fact that people in US cities dislike public transportation. In a previous job of mine, it was quicker to sit in 45 minutes of traffic outside of Washington DC then take the metro, and I was in an area with supposedly great access to public transit. Maybe if all these companies were serious about their workers' best interests they would do more to help invest in and lobby local governments to support public transportation.

> And many openly acknowledge that, but asking people to come to the office two times a week is not some sort of wage slavery.

If you were hired with the expectation to come into the office x number of days, that is fine. A lot of people were hired with the promise of being able to work remote, so changing that is where it tends to be a bit of a bad situation.

In general this animosity between office work and remote work advocates has gotten out of hand. These can co-exist.

[mehlmao]

I've worked with people (remotely) in the past that I'll still hop on Discord and play video games with.

Working remote lets me make connections and friendships where I live, and have more time to do so than when I was spending hours each day commuting.

[krona]

If that means only hiring at half the rate because of the longer onboarding ritual then that's a good a reason as any not to do remote onboarding for some companies.

[myth_drannon]

But Meta is famous for its "Sink or Swim" cut-throat culture.

[chii]

> you adjust as needed to support new employees

the adjustments may not be possible. People tend to build rapport easier when physically close by, and this rapport is how trust gets built up over time, and only by having this trust do knowledge and institutional culture gets transmitted.

I do believe that junior/new grads who come into a fully remote work place would have a harder time to build trust with existing employees, get less mentorship, and/or find it hard to "gel" properly.

[lopkeny12ko]

You cannot blame company policies for the poor performance or low motivation of individual junior developers. My team has hired multiple junior developers in the last 2 years, all of whom are 100% remote, and they've been doing excellent.

[zeroonetwothree]

I’m going to trust Meta’s sample size of probably thousands over your sample size of “multiple”.

I had six juniors join my team remotely and three are amazing, two are ok (do useful work but need handholding), but one is pretty bad. However I don’t know how that compares to non-remote hires so it could be that it’s worse.

[november84]

I totally get that, but it’s important to look who’s providing the information. I highly doubt the analysis was independent of bias.

We know many companies want to push for back to office work and commercial landlords are bag-holding hard…

Some food for thought.

[marcosdumay]

This is really not a matter of sample size.

If you want a real assessment of the problem, you will need to think about sampling, be careful about what you test, have a good amount of respect for the factors you don't know.

Nobody on this thread has done any of this, but one is a huge company messing with the lives of many people and claiming they know exactly what they are doing.

[Arkhaine_kupo]

I think they are not blaming individual junior devs, they are saying across their hires, they noticed a trend.

You said your team hired multiple junior devs, I am gonna assume its less than 10 hope that is fair assumption. Meta has hired almost 40,000 people since 2019. Even a conservative estimate of how many were engineers, and distribution of junior and senior roles. I think it is fair to say they probably had enough people to draw meaningful conclusions.

Perhaps individual teams, with careful hiring practices and team fit interviews can make sure juniors thrive even in a remote position. But on average, larger companies cannot be only staffed by super teams, and sometimes that means some people suffer to make sure the median employee has the best chance to succeed.

Its not too far from having junior limited to some staging environments instead of production. Some juniors might be ready for the big leagues, but giving the keys of prod to any junior will probably cause some headaches on companies with 80,000 employees like Meta.

[karmelapple]

+1 to this. Our small company hired our first full-time junior dev last year. They’ve been a fantastic addition to the team. We have also been doing 100% remote for over a decade, so our existing processes helped I imagine, too. But I’ve got another data point that a junior dev can certainly thrive doing 100% remote work.

[onlyrealcuzzo]

It's easy to assume other companies are incompetent, and your company is amazing.

But it's harder to onboard at a company like Facebook than at a startup. Every company could have better processes - but when it takes months for even top-tier talent to be able to do basically anything - it can be extremely helpful to be in person during that time...

And, sure, everyone is different. I'm sure some special snowflake will reply - but it isn't for me!! I'm speaking in general - as Facebook is...

[noisenotsignal]

Can we assume that Facebook did a good job onboarding remotely? I guess maybe the argument is that given poor onboarding, in-person is better than remote, and going back to in-person saves having to invest in better onboarding. But that’s just an argument that the status quo is easier - of course it’s easier to keep doing what you’re used to, and of course traditional onboarding is geared towards in-person! And that’s quite unfortunate because you do have companies onboarding remotely quite successfully so there’s possibly nothing inherent to remote onboarding that makes it strictly worse.

[myth_drannon]

Right, and it's up to the team to go above and beyond to help them. Managers and senior devs need to step up and get involved more with mentoring and helping.In the office you can gage stress levels of someone who is stuck,lost and not asking for help. In remote they will just hunker down and work nights to try to push through, it happens more to junior than to senior (usually).

[bproven]

yep - it is the culture that determines this. Ive worked at full remote companies that took time to mentor juniors and it worked fine

[paxys]

To me this sounds more like – engineers who have been at Meta < 3 years (so were hired remote) are performing worse than those who have worked there longer. Which...doesn't sound that crazy and doesn't have to be related to remote vs in-person at all. Of course your tenured employees are better at their jobs.

[oofta-boofta]

This was the most obtuse portion of the entire diatribe, and further cements my rather low opinion of Meta leadership; they have all the data in the world, but have no idea how to make proper conclusions from it.

[alfalfasprout]

Anecdotally... GOOG/Meta haven't updated their interview process significantly and it now just screens for quantity of leetcode grinded. This is likely a poor signal for actual engineering competence.

[lapcat]

I wonder how that analysis takes into account the general craziness and stressfulness of the pandemic, which was the cause of many employees joining Meta remotely? Not to mention, Meta did a crazy total amount of hiring during the pandemic, ramped up very quickly, which by itself can lead to inefficiency.

[guywithahat]

The irony is they’re tip-toeing around this but it’s been known remote workers don’t perform as well for as long as I’ve been in the industry. There’s a reason CEO’s will often come in touting how they’re going to promote wfh and then back off once they see the performance metrics their policy creates

[strus]

> it’s been known remote workers don’t perform as well for as long as I’ve been in the industry

I've worked in the industry for 10 years now, and I don't agree "it's been known". Any measurable source of such claims?

[wootland]

> it’s been known remote workers don’t perform as well for as long as I’ve been in the industry.

It most certainly is not "known". I've managed remote teams, I've managed in-office teams and I've managed the same people in both scenarios. At no time was the office advantageous to innovation. Quite the opposite, it's a soul suck and once you free people to work how they are comfortable, their creativity will skyrocket.

[EMM_386]

I'm 2 decades into this career with the last 10 being remote, and I am as productive as I was in the office.

It certainly takes discipline but I treat working from home the same as if I was in an office.

I am at my desk during business hours. People can Slack me, email me, call me on Teams, whatever. Besides, everything is logged. Every line of code, pull request, login time, email sent, all of it. They know what I'm up to.

[dongobread]

It does not surprise me at all that performance ratings are correlated with office presence, but not for the reasons they imply. The performance evaluation system at Meta is heavily biased towards political visibility and grandstanding and these are simply much easier to achieve in a physical workplace.

[zpeti]

I wonder if the eventual result of all this work from home debate will be that it's something that's earned after a period of time. e.g. 1 day a week after 6 months of employment, 2 from 1 year, etc.

That actually makes perfect sense to me from a practical perspective.

[wootland]

It doesn't make sense because the biggest advantage to WFH is that people can live in a lower cost of living city (and companies can hire the best talent globally). Forcing people into the office even once a month means you're tied to the real estate market next to the office.

[callahad]

Gradual phase-in makes sense to me, too, but who's going to be in the office to mentor those folks?

[zeroonetwothree]

More likely based on level than tenure

[agos]

I'm very curious as to what constitutes "performing better" in the context of this metric

[zeroonetwothree]

Selling more ads

[concordDance]

Matches my experience.

[wootland]

> our hypothesis is that it is still easier to build trust in person

Nothing builds trust like callously laying off tens of thousands of people over multiple rounds of layoffs! Want a trustworthy environment? Earn it. Really though when he says "build trust" what he means is that he (Zuck) does not trust his employees unless he can see them.

[andsoitis]

> he (Zuck) does not trust his employees unless he can see them.

You know that's not true. Facebook has offices around the world. No single person can "see" 80,000 people in any meaningful way.

[wootland]

His ideology and distrust is propagated down the managerial ranks. It's an entire culture built on distrust and exploitation.

[laweijfmvo]

Or maybe they lowered their hiring bar over the last few years, when employees had no choice to be in-office...