Google’s Scientific Approach to Work-Life Balance | Hacker News Reader

Google’s Scientific Approach to Work-Life Balance

blogs.hbr.org

93 pointskratiki49 comments

Is it possible for anyone who thinks for a living to truly 'segment'? The best I can hope is really intense distraction: mountain biking, sailing, even golf. It's amazing how little I consider anything else while playing golf--but I couldn't get away in the middle of the afternoon unless I had the flexibility to get good work done from 4-7AM.

Intuitively, I'd bet job satisfaction is right up there with mate selection as a happiness driver, so it's certainly worth understanding.

But thinking about 'work' (the task at hand) should be distinguished from thinking too much about all the merde that is 'the job.'

>"Is it possible for anyone who thinks for a living to truly 'segment'?"

Sure.

I'm finding this really hard to articulate, but here goes...

I feel like this should be self-evident, but we've been trained to think otherwise, to believe that giving so much of our time and energy is completely normal or even noble.

I've worked with my hands and with my mind, as a full-time employee, a union member and an at-will contractor.

There's actually nothing terribly special or unique about thinking for a living - it's still work with energy and opportunity cost for doing it. When the toll is primarily mental it's harder to perceive, but just as real.

The big difference between knowledge workers and everyone else is what individuals and workers as a group will put up with.

Tech workers accept being sent home with laptops and cellphones by an employer who doesn't want to pay three shifts worth of coverage for a 24/7 operation. Elsewhere this is prevented by unions, regulations and the implicit understanding that allowing the next man to devalue himself today will devalue the group tomorrow.

I have a difficult time stopping thinking; I consider this a big difference between thinking and acting: I would even argue I do not have conscious control over what I am thinking about... it isn't even clear to me I have conscious knowledge that I am thinking about something (hence why sometimes I seem to get more done while asleep than I do while actively thinking: sometimes I purposely go to sleep while working on a hard problem as there is a good chance I will wake up with an answer). So yes: I agree with you that thinking and acting can have similar costs and pains and should somehow be reglated to avoid serious problems, but I haven't figured out how yet: I don't even use email, and while I'm clearly addicted to my phone (I am at a party right now... clearly not ;P), taking my phone away would not allow me to stop thinking about the code I was working on earlier today... I was momentarily distracted while talking to some people, but as indicated by the person you were responding to, that's all I seem to be able to hope for: finding things that are sufficiently distracting to let me reclaim my own brain.

I know the problem, teaching can be a very absorbing occupation.

What I do is keep a notebook and write stuff down. That seems to 'earth' the thoughts in some way so I can stop thinking about that. I read the notes when starting a working day.

Surprised at downvote, just sharing my own experience.

> The big difference between knowledge workers and everyone else is what individuals and workers as a group will put up with.

There are exceptions, such as in Finland [1]

> Overtime work is allowed in Finland, and represents any work done extra the maximum regular working hours, to a maximum of 138 hours in a 4 months period and a total of 250 hours per a calendar year.

[1] http://www.workinfinland.com/what-are-the-normal-working-hou...

Sure, that's what the written rules say, but in reality, people put up with a lot more work. For example, they may not count work done outside the office, or even work done after 6pm in the office.

This is true, I suppose the difference here is that workers are offered some level of protection. I have had a Finnish person tell me he couldn't legally take on additional work before.

Why would you want to segment your work though?

Value creation is a major part of life, why try so hard to separate it?

If you have something fun to do, like going out, working on your hobbies / relationships / health, go ahead. But you don't have to put those activities in two "time zones". It can be a homogeneous mixture.

* Elsewhere this is prevented by unions, regulations and the implicit understanding that allowing the next man to devalue himself today will devalue the group tomorrow.

yeah, but we make 3x as much as the union protected workers. It's union protection that ruining this country, look at the entitlement problem, hell, you only have to look at the train driver in chicago who fell asleep at the wheel, it a non union protected area she'd have been fired, instead she goes right back to work and drives a train up an escalator asleep again.

I achieve my segmentation by studying, and I've been very successful at forgetting about my job when I'm not in the office. Not everyone's cup of tea, though. Other people I know achieve it with sports (not my cup of tea, at least not competitive sports), and considering what we know about how it affects body chemistry, it's probably a very good idea if you're into it.

Job satisfaction is, not coincidentally, a lot like money satisfaction. If you're not worried and/or pressured, you can pretty much forget about it; but if it's a problem, it may eat your life pretty quickly.

In any case, just remember: the most important resource by far is time, not money. Money can be the most urgent resource to attend, at some shitty/unlucky times; but in the long run, aim for the highest amount of time for yourself. I'll add this: the worst job I've ever had was also the one that paid the most.

Sample size of one, but yes, it is: Years ago I took my work stresses to bed, each and every night. Eventually I realized these stresses were feeding me to a vicious circle: Stressed in bed, sleep poorly, fail to refresh, get less done - and do stupider things, get more stressed.

Eventually I just flipped the switched and stopped it. I wish I could recall the moment (or period?) more clearly so that I could describe what I did well enough to make you believe you can do it too, but it was a long time ago, and I stopped worrying about it pretty soon thereafter.

Will those stressors be there in the morning? Is there anything you can do about them in bed? Stop worrying.

(And stop convincing yourself there is something you can do about them in bed, because you know very well that you need a fresh mind and a fresh perspective to do your best work. And you know very well you are not going to get those things worrying with your head on the pillow.)

Rituals can help, such as commutes: Leave the office stressed, put things out of your mind as you go home. Be mindful of the sky, the clouds, the people around you. Make up stories about them.

On the way in to the office, plan how you will evaluate the stressors.

And use the "will ignoring this get me fired" rule for everything. That really helps with focus.

(I still think for a living - but now I'm quarter-owner of a consulting firm - stressors now are ever so much greater, but my stress is generally less, because I segment and defer.)

One thing we know from studies is that when we effectively delegate work to someone else we trust, we no longer worry about it. The other thing we know from studies is that inside our brain we consider the 'future self' as another person - fMRI studies have shown that 'future self' is the same parts of the brain as we use to think about OTHER people.

The result is if we trust the future self, we can effectively delegate to someone else - ourselves!

I think when you figured out 'the switch' this is what you are doing.

One exercise is to say to yourself "This is a problem for Future <insert your name here>". This is surprisingly effective! Of course this only works if you trust your future self's ability to remember... an organization system really helps here. It doesnt matter WHAT system, just that you know it will work.

Intriguing observation. I hate to use this term, but it "resonates" with me. I'll mull upon this awhile....

EDIT: I just realized that I do exactly this when the "usual tricks" don't work, invoking "tomorrow me" as needed. Talk about context-dependent recall - I don't think I've ever thought about that technique other than when I am using it. Thanks!

Software engineers (and many other information workers) are lying to themselves if they think this is even possible (to segment).

Proof: a large number of breakthrough designs / problem solving occurs when I am in the shower, on the toilet, or lying in bed going to sleep... it has long been proven that information workers are solving problems 24/7 365.25 days a year - you are never paid 'by the hour' for your work - your brain is solving problems 24/7 non-stop.

Indeed, this is precisely why allowing staff to work from home and minimizing enforced 'attendance hours' works.

You want to blur the lines between work and life... quite simply because they are already blurred. It's like the infamous 80s/90s Waterfall SDLC - it's like trying to make the waterfall run uphill (all projects end up Agile in the end). So why fight the inevitable - you are only stressing yourself by doing that (as proven by the evidence in this article - the 'segmentation seekers' are trying to achieve the impossible!) - and it is medically proven that stress causes cancer, diabetes, heart disease... and death (2009 Nobel Prize in Medicine). Corollary to that: blur all of life and live healthy and happy, or attempt to segment and suffer stress and die painfully.

I don't see why not. When I clock out of work, it all stops. I don't get work email on my personal phone, I sign off IM, etc. I am completely unreachable (except for calls, which nobody ever really does). No matter how stressful the day was, or how much stress is coming up, there's a clear delineation between work and life, and one doesn't bleed into the other.

It has kept me pretty happy with my job for more than a year now, even though I've taken almost no vacation (which is not as bad as it sounds, I work remotely).

Both of these positions on the subject are needlessly binary. I often find it the case that, even though I erect careful boundaries between work and non-work, when I leave a problem on my desk, I will continue to ruminate about a solution throughout the weekend, say. Similarly, I will Strategically avoid email when I'm recharging. I don't think it's so easy to boil down as black and white.

Sure, but that's the fun part. There are exceptions, obviously, just don't make them the rule.

This is probably why Intelligence analysts steadily rank as highest in job satisfaction [1], you literally can't take your work home with you.

[1] http://money.cnn.com/magazines/moneymag/best-jobs/2011/caree...

I use physical activities as a counterpoint to job-thinking. Fixing an engine or cutting down a tree requires attention and engagement, but aren't mentally taxing, so it's a good way to unplug. Plus, spending a weekend actually doing things instead of pushing buttons and flipping bits is very satisfying.

I think about work in the shower, before bed, etc. But watching "Frozen" with my daughter, I totally forget about work. Or when my wife unceremoniously hands me the baby with a "I already changed a poop diaper today, your turn."

You mean you don't 'consciously' think about work... your brain is actually still joining dots in the background 24/7... unless you stop it by using drugs!

These Boss 3.0 names obscure the old concepts. Google's moves are part of building a "company town." [1] People Operations is "human resources" [2], and its People Science is the same "scientific management" data-gathering used by slave-owners [3], corporations and the military.

Model View Culture minces fewer words about how tech corporations mold their workers' individual identities: (http://modelviewculture.com/pieces/tech-workers-political-sp...)

[1] http://www.nytimes.com/2010/02/19/us/19sfvalley.html?pagewan...

[2] http://www.google.com/about/careers/teams/people-operations/

[3] http://hbr.org/2013/09/plantations-practiced-modern-manageme...

I'm interested in identifying companies that focus on results and not hours of "butt in seat" time. What's the best way to do that?

I prefer to be a segmentor, but it seems like a lot of companies prefer integrators, to the extent that they want, as alluded to in another comment, to eliminate work/life balance by making work and life synonymous. That's definitely not for me.

This can depend more on the particular product group and manager you work for than which company. All big companies have less than perfect managers, and some people have a hard time accepting work/life balance choices that are different than their own.

While I applaud the effort, the only issue here is the sample is restricted only to "Googlers." That's a bit of bias for their sample already. Google is hardly an average company, so the results they glean from their employees might not generalize as well.

Interesting stuff. I feel like over the years I've moved from being an integrator to segmentor. I think it just comes with experience. I have a better understanding of what to expect, which "emergencies" I can ignore or postpone. How to avoid putting myself in a potentially stressful situation. I have a metric for this too. I suffer from a mild medical condition that flares up when I'm stressed. I used to suffer from it at least once a month. Now I'm down to every 6 months. I credit maturity.

I wonder what sort of life experiences might influence whether one is a segmentor or an integrator. I am heavily a segmentor, which I suppose I'd attribute to growing up with divorced parents. I was shared between my parents, one week at moms and one week at dads. Each house had it's own rules, toys, friends, etc. I'd never considered it before, but, it seems like I was raised to be a segmentor.

I wonder how they control for the Hawthorn effect in their 'fix it for just some people' experiments.

Heh. I took a photo of that same girl(I think) on the cover photo when I was in India: http://www.flickr.com/photos/miloshadzic/7102823323/

Nice, that was shot on film right? Love the look. Also, seems like a bit of a light leak... a tip: I once glued some black foam-paper on the door-hinge to my old film camera, because the original one from the 70's had worn off haha.

Yeah all of them were shot on film. I had two cameras, one was a Yashica GTN(those that don't have a leak) and a pocket Fujica. I'd had fixed the leak if I knew about it but the thing is I got it right before the trip heh.

I would be curious if Googlers on this list feel that Google does a good job putting these insights into practice.

I didn't really get the point of the article, so I'm just going to comment on work-life balance at Google in general.

Basically, you are freed from any unnecessary constraints in choosing your own work life balance. You are judged by the work you do, regardless of how long it took you to do that work, or which hours of the day you did that work.

The company goes to reasonable lengths to allow remote working, but they can't make up for the inherent disadvantage of not being in the office/campus when other people are.

In my opinion, that is the best that a company can do in terms of promoting work life balance. A company could also actively try and stop people from working more than X hours, but I don't see why that is a good thing, if a person really wants to do it.

Many posts on HN suggest that when a person chooses to work long hours, this is (A) an irrational choice on their part and (B) imposes externalities on other people, and therefore should be culturally discouraged. On (A), it's hard to judge all cases, but some people really want/need the money. On (B) that is really just wanting to avoid competition. If another person is willing to work longer hours or accept less pay, that is going to harm me, but it is also how the free market works. I'm happy to accept market wages because that's how capitalist societies work, and both in theory and practice, they work* pretty well.

* And by "work" I mean that they provided the overall best outcome, including for the worst off. I added this disclaimer because there is always one comment that says something like "yes capitalism works... for the rich".

Except for B) having nothing to do with free market. 1)The problem with a long-hours work culture is that long hours don't translate to productivity, and it's rather hard to quantify productivity in a programming setting. It's an emotional/managerial phenomenon.2)It's not avoiding competition either. If the optimal work hours for a programmer are 4-6 hours, then 10 hours surely will lower his/her productivity and just project "more work". As a result, we get a badly competitive micro climate based on "projections" of more work.

I suggest you look at how Finland approaches non-competitive study in elementary and high schools, which propelled them to the top of the world in terms of quality.

I'm not describing encouraging people to work long hours for the sake of it, but rather people working longer hours in order to get more done. It seems like you don't believe it is possible for a company to reward people based on their actual results, ("it's rather hard to quantify productivity in a programming setting") but that is what Google does. No one really notices how much time you spend in the office.

So I would accept your points (1) and (2) to the extent that they apply in a particular setting. But these arguments don't apply to Google. If a person at Google chose to work longer hours, and ended up being less productive, that would really only harm themselves.

1) No such thing as long hours for the sake of it. It's always to get more "done". And like I mentioned in my parent post, there's no such thing after a certain number of hours. 2)I don't believe it's possible for a company to do that, or at least not to the extent that you're saying. 3)Based on some of the interviews people have had here with Google, it seems that Google could use some more "reward based on performance" in their culture.(or a lot)

p.s. Keep in mind that my competition point still stands. Longer and longer hours are of no good. It's a slippery slope. And yes, I think such competition is poisonous and detrimental, even if some more work gets done. Working more hours just pulls everyone down,* in regards to your free market comment. It's price competition.

*even if it's not projection, but actual performance. Soon enough, everyone is working more and more to catch up to you, and everyone's miserable. That's not sustainable practice.

Soon enough, everyone is working more and more to catch up to you, and everyone's miserable.

That's an argument against high performance regardless of hours. But I doubt many Googlers are miserable because they have to "compete" with Jeff Dean.

I guess it is. It sounds wrong to me as well, but there's only so much performance you can fit into certain hours. Thus, more hours are bound to become the norm, in my opinion.

That assumes productivity remains constant, which has historically been a poor assumption, both in the economy at large and with individual people. When you do something a lot and have a modicum of intelligence (and most Googlers do), you figure out new, more efficient ways of doing things. That lets you accomplish more while working shorter hours.

The unfortunate thing is that if you're really good at making things more efficient, people want you to do it all the time, but if you're paid for results (Googlers are) and decent at setting boundaries (by and large Googlers aren't, but this is an individual-responsibility thing), you can work out compromises that give you both professional advancement and time for a life.

>>>The problem with a long-hours work culture is that long hours don't translate to productivity,

I have no idea how you can make such statements for everyone. Feel free to talk about yourself. It is ridiculous to claim that working "extra" does not result in anything "extra" for everyone. That may be true for you. It is in fact not true for me. Now you can call me deluded. But I think I am smart enough to measure my results and decide accordingly.

he problem with a long-hours work culture is that long hours don't translate to productivity,

[Citation needed]. Note that citing a source that vaguely asserts data exists is insufficient. So is citing a source showing that mean hourly productivity in construction/manufacturing goes down.

Make sure to differentiate between lowering productivity (=production/hours) and lowering product in your answer.

I think it would be disingenuous to consider two years worth of studies "insights"; I think the truly interesting stuff will likely come out of clustering machine-learning algorithms applied to the data ten, maybe twenty years down the line (when some Googlers' children have grown up and we can see the effect or lack thereof on them as well as on the Googlers themselves).

I get the impression that Google errs on the side of "provide more information"; for example, food in the micro-kitchens has been labelled with traffic-light colours for many years now. Many of the interns I met at my last internship there went out of their way to eat more of the food that was labelled "red" because "that clearly means it tastes better". Whether that made the life of the interns in question "better" or depends entirely on the metric you choose to measure with.

No idea as like others have said the survey is only 2 years old.

What I can say is Google is the first company I've personally worked at where it was easy to always work. At all my previous jobs (video games) the dev kits could not be taken home and even if there happened to be a PC version there's no way I could have taken home the terabytes of source data needed to work on the game. Nor would any company have let me if I could.

The short of that meant once I left work I couldn't keep working.

At Google on the other hand my work email went up 10-100 fold over previous companies. I could read that anytime day or night and even if I didn't intend to read it at home I'd open up my laptop and given that I left gmail running I'd see new messages, open a few without thinking and get stucked in and 20-60 mins would go by at ~12:00am.

I saw several other employees responding to email between 11pm and 1am all from home.

I'm not saying that's bad for work/life balance. I have no idea. I certainly felt good to respond to people as soon as I noticed but maybe it would have been better to disconnect?

Similarly I could ssh/vnc in if I wanted to. And, working on Chrome which is open source meant I could work anywhere on any machine I chose. Something I wasn't used to from previous jobs. My understanding is some of those things are far more common for other programming jobs. My financial friends all have ways to log into work from home when they need to.

Yet another is being involved in a web browser meant it was easy to get sucked into tons of mailing lists. The WebGL list, the Web Audio list, the W3 lists, the Web Apps List. All the places where browser standards are discussed and that discussion goes on 24/7. It's very tempting to keep checking how people are responding to your comments or where some other part of the standards discussion is going. So that's also hard to turn off.

Again I don't know if that's good or bad for work/life balance. It certainly feels good to participate. Is that work? Or is it more like going to a PTA meeting or city council meeting?

No where was I encouraged to work after hours by Google. It was completely up to me to decide if I checked email at home or if I worked on stuff at home and if I stayed late or went home early. All I'm saying is it was easy to do these things, something I wasn't used to from previous jobs.

PS: You don't have to work at Google to participate in those mailing lists. If you want to influence web standards just get involved.

Let me help you.

> Again I don't know if that's good or bad for work/life balance.

It's bad.

What's missing from this is that Google's culture is designed to eliminate work-life balance by making as much of life as possible focus around work. A significant part of the compensation for Google employees is in the form of good "free" meals (including dinners), lots of services, entertainment, etc., all without leaving the Google campus. Would be interesting to see how the results are different from similar surveys done at a company like Microsoft that doesn't have such an all-encompassing work environment.

Microsoft pretty much invented the all-encompassing work environment. The nickname in the 90s for the Redmond campus was "The Velvet Sweatshop".

Microserfs

Blah-blah-blah HR stuff, blah-blah-blah, leaders, goals, performance, roles.

The true science approach should aware that massive surveys can measure everything but the thing you want.

People don't "trust" that survey anonymous and trying to reflect what they expected to say in survey. Human can't seriously tell something to the stone, communication doesn't work this way. If there is a really stone of anonymous survey, human have to give him imaginary human trait, expectations and ability to evaluate him. "What are you want from me, dear stone?" So, confidentional procedure doesn't mean that people anser like it is confidentional.

Interviewer presenation - that the sociological surveys is actually measure, and there is only the ghostly shape of personal realtion behind this.

The second thing is the question interpretation: "I don’t like to have to think about work while I am at home." Can mean: 1. I don't like any serious thoughts at home. 2. I thinking all time about work ant don't like this. 3. I just don't like and not thinking. 4. I'm deep in self-estrangement and prefer not to thing about work at all.

Event "think" can be perceived in many ways: specific persons, tasks, it domain at all.

How massive survey is doesn't matter until we have raw data in hands. Some manipulation with R or whatever and there is a some groups and strict classification, some more and there is completely different classification, one question data can change everything or be just excluded to make picture "clear".

The point is that any changes in methodology of survey researches usually demonstrate applied catastrophe theory when any shift can significally change the results. Don't trust any conclusions without detailed step-by-step description of methodology.

In general, measure the pressure, scan brains, make tests aimed to somatic responces, use anal probes, but dont't ask the questions if you really want to understand something about people.

Reply on news.ycombinator.com