The productivity claim has been removed from the Japanese report. An erratum at the bottom of the page dated November 8, 2019 says this:
"In the announcement dated October 31, one of the listed "improvements" from the 2019 Summer Work-Life Choice Challenge was an increase of 39.9% in labor productivity (sales revenue per employee) in August 2019 compared to August 2018, with a graph below.
"While this number is factual, it is not solely the result of this challenge, and was achieved due to a number of different factors.
"To avoid misunderstanding, we have removed that claim from the above summary of the direct effects of the challenge."
Especially the definition of productivity as "labor productivity (sales revenue per employee)" is interesting and adds some perspective to this claim.
It's interesting that working on a two plus year product cycle does not add to productivity, at least not how it's used here.
So as an employee it would seem to be a better career move to always work on something that can be shipped within a year.
If you havent already done so, go and change your entire organizations default meeting length to 30 minutes. Force people to justify why they need an hour. Still not sure why it isnt the default.
People get to be human and do things like use the restroom, get a drink, or just take a breath in between meetings without being flustered and apologizing for being late.
Just sharing my POV. You do you. :)
I've only been to one company that had a sane meeting culture. Along with every meeting invitation there were attached documents that you were expected to read. You were expected to have notes and comments ready. The meetings mostly were not supposed to be a time where you worked vague things out.
Specifics are already easy to discuss async.
My personal observation is the timing matters less than the number of participants. More people usually equals less insights and more wasted time. One rule could be "if you can't emerge with an action item you shouldn't be here." But I'm not sure if that's practical as an organization grows and social/political relationships matter more.
“An academic quarter (localized into various languages in the countries where it is practiced[a]) is the quarter-hour (15 minute) discrepancy between the defined start time for a lecture or lesson ("per schema") and the actual starting time, at some universities in Europe including Austria, Belgium, Croatia, Czechia, Denmark, Estonia, Finland, Germany, Greece, Hungary, Italy, Latvia, Netherlands, Norway, Poland, Portugal, Romania, Serbia, Slovenia, Sweden, Switzerland, the UK and the United States.”
However, while meetings were sometimes explicitly scheduled for a quarter past, if your thesis supervisor schedules a meeting at, e.g., 9 AM, then be sure not to show up at 9:15. :)
Complete disrespect for him, absolutely the norm for me :D He learned to be less angry and I learned to be less "late"
It works really well, except for non-classes no one is ever sure if you were told UofT-time or normal-time. For example if you agree to meet with a prof or ta outside of class..
I think I would have preferred if they just always scheduled things at "xx:10 to yy:00" instead of having the ":10" as implicit.
A lot of these organizations put status updates as meetings though, which is never necessary. There are a lot of better ways to send status updates.
Force people to justify why they need a meeting as well, have an agenda, keep a tight schedule, don't let people change the agenda or rant on. One thing I found in the scrum style meetings is that people love to be on the soapbox for their allotted time.
* a firm that was demanding devs return to office already in January pre-vaccine, justifying it by paying for Ubers & doing PCR testing in office
* a firm which in its job-spec explicitly listing that "8a-6p are standard hours" and "plus other hours as required for role / due to support / meetings out of region / some flexibility required"
what a shitshow tech can be
So why would that be a "bad word around these parts" ?
Unions gave us the weekend, can’t be all bad.
Some revolving list of "other team __" being slow/unreliable/hard to manage/etc..
The same 1 or 2 senior devs is tasked with taking away some discussion points to have with them.
The meetings would only take a few minutes with Kanban, assuming the backlog stayed sorted. I just asked where the new tickets had to go in the current sorted list.
These meetings would be much faster if project managers didn't clog the backlog with stream-of-consciousness tickets.
However I was never an Agile guy. I trimmed the process heavily, and took up most of the backlog maintenance so that others could work.
Our product manager sits in on the meetings such that he can says yes/no when questioned on tickets, but for all I know could be watching soap operas given his level of engagement.
So the best approach would probably be to schedule 30min meetings that are always followed by 30min free time; so that everyone knows there is an urgency to be brief, but it's not the end of the world if you overrun 10mins. Otherwise you end up with back-to-back meetings and "I have to go, sorry", even if we're not actually done, or the 1h snoozefests.
I think there is space for some software here, that could automatically tweak one's calendar to "block out" post-meeting time by a predetermined amount in Outlook (yes, don't try to pry people out of Outlook, you're not going to win).
> meetings that are always followed by 30min free time
It seems better to have no free time between meetings. That way everybody is much more motivated to keep them to the allotted time. 30 min is much too short an interval to do serious technical work.
It absolutely blows my mind that 99% of office roles are still 5 days / week, Monday to Friday - why is there basically no variation on this model?
It annoys me so much that I recently launched https://4dayweek.io/ (Software Jobs with a better work / life balance)
For one, there was already a massive shift that nobody noticed. A few decades ago people complained about how awful and soul-crushing 9 to 5s were. Well, someone decided to reform that. Now 9 to 6 and 8 to 5 are everywhere. I've even known some people who get two 5 minute breaks at work, and now their job runs from 9 to 6:10.
Or no lunch break ?
You do however get RTT (basically paid holidays) to get back the hours that you do above 35h (usually half a day a week to compensate for the 40h classical work week)
I worked for an organization where people had one flex day off every two weeks. In practice, it made Tue-Thur the only days for collaboration as it was otherwise normal for someone to be out of the office. It worked because it was government with long project cycles. It would not work for my current company. It would cause issues with clients and project speed.
Especially when you run 24/7.
Then you know it's really bad when a 24/7 operation "rises" to a challenge by calling all hands on deck. And at the same time battle stations because they don't even know the difference.
Collaboration is often limited by the pace of a human interaction: a design review, a question about some code etc. The only way to not have this speed limit is to remove all such interactions which is very difficult (and probably not desirable either).
If I worked on off dates and my colleagues on even dates, the shortest time for me to get a code review would be 24h! Most communication can (and often should) be async, but there is a difference between a 10 minute turnaround and a day. Also, if something is very urgent one should be able to get hold of someone e.g give them a call. Developers hate being interrupted, for good reason, but that doesn’t mean developers should never expect to be interrupted.
I think a shorter and flexible workweek is a good idea but having time zone overlap or “required work days” to maximize overlapping work time is probably required in most cases.
Consistency is good.
That's much different than a pet project that's being given away for free.
Also remember that opens source includes things like the Linux kernel. That "pet" project underpins almost all software that the hypothetical corporate uses and makes, and does it without the meetings and the deadlines.
And if we are talking about some of the most important software projects, approx. zero of them are run by hobbyists. Some niche aplications - sure. But not the most important pieces.
(Disclosure: for the last few years I employ 4-8 people to work on GNU GPL licensed products, so I know a thing or two about how it works)
I’ve also developed and led teams that developed many OSS projects and very successful consumer applications. Not once did I micromanage my team’s working hours. They make their own schedules as people know best when they can produce good output.
As for 'micromanaging working hours', it's not a question of micromanagement, but of communication. It is far easier for everyone to work in the same rythm, knowing that you aren't distracting a person you need to contact from his private life.
That’s not flex time at all, that’s a fixed schedule where everyone has either even or odd Fridays off.
If your coverage plan relies on the customer knowing who to call when, your problem isn’t flex time.
It's a fairly small number of companies atm but it's growing fast
I remember my father telling me about a push to a 4 day work week in the 80's at local 6 in San Francisco.
For some reason, the union officials didn't want it. Looking back the contractors probally nixed it?)
I remember how sad my father was over that. He was getting older, and wanted a job easier on his body, and he thought the 4 day work week would be enough time to start a side business.
I did electrical for awhile, and mirrored his thinking. I used to think 10 hr days were not a problem. (Union construction. Mon-union construction is way harder work.)
There is a definite drop off point where a technician/mechanic/welder isn’t working at even 80% normal capacity, and that’s where mistakes begin to happen. And unlike software, mistakes can be costly to undo or perhaps even impossible.
The flipside of course is that their hourly wages must go up to match the decrease in hours worked. And overtime pay is a big motivator for many skilled manual labor jobs.
The most productive cadence over a season was 5-1 with 9 hour days. 4-1 at ten hour days was a close second. We would switch to 4-1 when drives got longer.
It felt like a sport more than a manual labor job. Your trying everything you can to maximize your tally.
Also after time wouldnt there be inprovement due to experience? (Or is it cancelled out by injury?)
Two or three years after it is checked again and will be scheduled for fill planting if it doesn't have good enough density.
You improve your speed a lot at first. 100 trees your first day, 1000 by the end of your tenth day. Meanwhile vets are putting in four thousand on the same ground at 12 cents a tree. Definitely pay your dues and a company won't pay you minimum wage top ups for very long.
It takes about five or six seasons to reach your peak in terms of knowledge and skill. Some would argue longer. There's efficiency of movement, controlling your emotions and mental aspects, reading the ground and spacing in triangles instead of straight lines. Not many planters get to triangles. There's also different ground to learn. Trenches in Alberta vs steep snarpy old growth blocks on the coast are completely different worlds. Coastal planting is hard to get hired for, mostly older lifers, interior you'd find many students on crews.
Everyone gets minor injuries and sore nagging pains that steadily get worse. It's often just a matter of how many days you last before you quit. Seasons can last from twenty to a hundred days, probably fifty is the most common. Being in good shape or even a little pudgy at the start helps but it's more a matter of pain tolerance and being able to set those thoughts aside and tune it out until you are numb to it. I came back to the truck one time with a three inch piece of stick jammed an inch in my shin. My pants were ripped, leg all bloody, didn't even notice it.
Like coding there's 10x planters too. They are called high ballers. Typically they are more fit but they still come in all shapes and sizes and male and female too. A top planter can make almost twenty grand CAD in a month, reliably 10-15k.
It's a great summer job, you meet genuine people, you learn a lot about yourself. It can fund a lot of travel or other adventure the rest of the year if you play it right. And there's others that would say it was the worst x weeks of their life.
However, it is 4 10 hour shits and not a more reasonable 4-8 hours.
I had enough energy to do a week’s work in four days, plus people got so used to me not being around on Friday that I pretty much got the rest of the year free of Friday meetings as well.
(I also just came back from two week’s worth of bank holidays on Thursdays and Fridays, which seems to be having the same effect).
The business owner isn't more productive at 5 days vs 4 but they make up for it by working 16 hours and 7 days a week.
It's not too dissimilar from working in a very early stage startup as a very experienced engineer: you know that there are lots to do, from infrastructure to the product code. You need to get your hands dirty all the time, playing multiple positions, to beat time to market.
As an employee of an organisation that you have no other vested interest apart from your salary? Yeah, this stress will only grind you, not motivate you, at least it did to me, the thought of "why am I sacrificing my life for a paycheck so someone else's grand vision [and ultimately, profits] come to them?".
So yeah, as an employee I'm much more productive and creative to my employer if I have time for myself out of work and good rest. As a business owner: you are running the risk all the time, some people cope with it by overworking to not feel like they are "falling behind".
I also run my own business and have never once felt the need to work 40 hours in a single week.
It’s your business so you can design it to work any way you like.
* Cultural norms.
* Poorer pre-existing efficiency.
* Introduction of new focus on time and efficiency creating optimization which fades over time.
* Gains on shorter-term metrics offset by longer-term, hidden costs (eg sufficient 'slack time' is asserted to sometimes enable innovation or tasks which management may prioritize lower but which have yield important benefits).
Over several decades I've been either an early employee or founder in four startups where I was able to help imagine, specify and implement what eventually became high-performing organizations. I'm often asked about "secrets of success" but all four had substantial enough differences that the only constant is they each were either created or evolved to have uniquely effective set of parameters including structure, style, culture, personnel, etc. Each org was uniquely adapted to it's competencies, customers, context and even era. Obviously, "good people" are "good" but figuring out what that looks like in advance for any particular problem set is error-prone and rife with confirmation bias upon post-hoc rationalization.
The best meta-advice I'm confident enough in to share are things like adaptability, the right metrics, perceptive leaders willing to question assumptions and change course based on new evidence. Unfortunately, these come across as vague platitudes. As near as I can tell, the magic is more in custom fitting a sufficiently winning mix of org parameters, from team to metrics to work hours and more, to succeed against the specific challenges and context the org must solve to succeed.
In light of my world-view that dozens or hundreds of interacting parameters need to fit (and that there are many workable "solves"), I feel that this experiment, at best, exposes another possible lever to put into your "solve" that might be beneficial contingent on a bunch of other factors.
Since leaving my last job, I've been working seven days a week, but at my pace. Lots of breaks and downtime (naps, even), and I don't mind taking full days, every now and then.
Amazingly, I get more done, in less time, than I ever did before.
In Japan, the traditional salaryman workweek is 6 days, with Saturday being "casual," and sometimes, a "half day" (only 6-8 hours).
We'll have to see how this works out. Lots of cultural hysteresis.
The traditional office layout is also totally open-plan with no sound deadening and your manager staring at you from the side. Of course, I'm not sure there's much expectation to get any work done, but I always wondered if people too picky for cubicles knew what other countries were like.
I do not consider Japanese business to be very productive. An engineer I worked with explained his extreme hours to me, by saying that during the "workday," he was in meetings the whole day, and needed the "after hours" to get tech work done.
Sounds about right.
It seems it's starting to change. Currently I'm leaving the company which indeed has 6 days work week (with the 6 hr Saturday) and literally every opening I'm seeing right now is 5d.
Maybe I'm biased though as I'm looking for jobs in IT companies or departments.
I don’t miss those cramped conference rooms, one tiny bit.
But they run their meetings in a very organized fashion. Everyone has their marching orders, when the meeting is done.
Japanese corporations tend to use a consensus-based system, where meetings are confirmation of consensus. Being at a meeting, meant that you were part of the stakeholders, and also part of the accountable parties.
Many times, people sleep in meetings. Their presence is all that is required.
We see reports like this all the time but barely any companies actually have the balls to go through with it long term.
210 comments: https://news.ycombinator.com/item?id=21433710
242 comments: https://news.ycombinator.com/item?id=21441689
In the announcement dated October 31, one of the listed "improvements" from the 2019 Summer Work-Life Choice Challenge was an increase of 39.9% in labor productivity (sales revenue per employee) in August 2019 compared to August 2018, with a graph below.
While this number is factual, it is not solely the result of this challenge, and was achieved due to a number of different factors.
To avoid misunderstanding, we have removed that claim from the above summary of the direct effects of the challenge.
I expect this works great for 2 to 3 months when it is novel and everyone is keen to prove it works, but will begin to regress back after that. Where it ends up longterm is the big deal.
https://www.theguardian.com/world/2021/mar/15/spain-to-launc... (March 2021: Spain to launch trial of four-day working week)
Like if you reduce everyone to 10h per week. Sales may go down from 4M to 3M but because hours went 4x less you can still claim 300% increase in productivity. It doesn't meant anything in term of efficiency, creativity, or actual performance of the organization. I would be interested to see actual studies on this.
How is that not a measure of productivity?
In software it is much easier to scale sales: the infrastructure costs when selling 1M copies and 1.4M copies are quite similar. Maybe some devops will have to spin more servers.
But programmers allready did their job.
Compare this to selling physical objects. If you can make 1M cars per month, you cannot easily make 1.4M per month: it requires more workers, machines, supply. Those are quite neglible in software.
Also your workers can make cars 2x faster than everyone else (better productivity) but nobody will buy them, since they are poor quality
What on earth are they printing at Microsoft?
> Each year the amount of wood harvested from U.S. forests is much less than annual forest growth. Land covered by forests in the United States increased by 4.5 percent between 1997 and 2012, even as suburban development expanded.
I don't think anyone 'could use more' paper work.
Actual result: everyone began printing screenshots from the system and storing these in file cabinets.
Typical work week 6 x 5 = 30hrs.
8x4 work week = 32 hrs.
Longer periods for developers to hit technical strides or remain in flow.
I had the deal with my boss that I will come in on Friday, if there was some important work left undone. I never had to come in on Friday :D
p.s. that phenomena is called "Parkinson's Law" i.e. work expands to fill the time allotted to it
Also if you want to have some "life" during week (gym? hobbies?) it is problematic.
I dont say that this is bad, but one of the things that works for young, single people.
I had to work 5x10 (then add 1 more for travel..) and even basic stuff like buying and cooking food could become a problem, especially if some day was 12h.
Scrolling through the Google Translated version of the original Japanese article, it seems a bunch of other changes were implemented at the same time, including more remote meetings, pushing text chat over meetings, and paying back more expenses. Also the number of printed pages just followed a trend that had been going on for several years, and the electricity usage was also already decreasing .
Let's say the typical job has 3w vacation:
40hr x (52w - 3w vacation) => 1960hr/yr
Now to get back to that same territory:
50hr x (52w - 12w vacation) => 2000hr/yr
I have not worked in France but as far as I understand, if you work more than 35 hours a week you will get the extra hours as additional time towards your annual leave:
So this isn't surprising. It would be interesting to know if those 4 days are filled with similar hours to a normal working week in western countries because of the overtime culture.
But, more likely, there are a host of well-known reasons that corporations don't operate like that. Primarily based around the fact that corporations don't make decisions, people in corporations do. And their interests may conflict.
Take the following quote from the article. It's laughable.
> The question was straightforward, as Schawbel recalls: "If your pay is constant, how many days a week do you want to work?" One of the potential replies to that question was simply, "None." But only 4% of the workers chose that answer.
"Raw output" by which I mean things on the level of "code written/tickets resolved" probably scales pretty directly with undisturbed hours of work, which WFH isolation + no commute is great for. So I can see that type of productivity going up.
However there's a short term trap involved in that - since by optimizing for the above, you are neglecting the ability to get stuff done as an organization - ie benefit from everyone's insight, have things properly stress-tested and socialized, planning company-wide programs, etc.
The second bucket of things benefits from activities that hurt the first bucket (namely, in person meetings.)
Now, it's possible that most of the company is in role that are measured for #1 only, however #2 matters to the overall health of the company and long term excellence. So it may seem weird, if one is accountable for churning out tickets, that they are being asked to work in a way that's not optimized for that, but the answer may be - this enables other work to happen in the company which is required for your work to be meaningful.
Unless they're running a good ol boys network, I really dont see that as a valid concern. They certainly dont care about us lower people's insight and the important people socialize things virtually.
Objectively not true. Say hi to someone at the coffee machine, chat, what do you do, etc - learn he's worked on something similar to what I am trying to do now - get some help.
If not for coffee-machine run-in, wouldn't know he exists, wouldn't know he's done that thing. Multiplied by a hundred micro-interactions like this a day.
Zoom/slack is less serendipitous, more planned. But if I don't know I need to meet you ahead of time, I won't.
Would you be better off in support forums for your specific issue.. reading about how others solved similiar issues? or asking a collection of people around the world going through similiar problems?
Talking to a 100 people a day for 5 minutes would be 500 minutes a day. Hard to get any work in.
Japan is noted for having to be "at work doing nothing" for obscene number of hours because the boss is also "at work doing nothing" and then having to go drinking afterward.
This is a recipe for unproductiveness.
Being able to actually have the day off because you don't have to emulate the boss breaks this significantly and is likely to have an outsized improvement on productivity.
In terms of work-life balance and job satisfaction, Japan does not rank very high. Not every country is in that situation.
So an initiative like having a 4-day workweek will likely be an improvement given those conditions.
If your company continues to make more profit while you work less it’s in your interest to not lose your wage position. It would be stupid to work 4 days to make more money for the company and make less than if you worked 5…
Everyone knows the ‘eight hour’ day was invented and has no particular meaning. So—make a human decision, and stop pretending you can science your way to something else.
Vaccines, nah? You get yelled at for that.
More work life balance? It's okay not to trust science.
I take it you don't have children, elderly or disabled family, other responsibilities outside work, or even hobbies?
Having a day free to go shopping, take a hike, do laundry, and even just not have to worry about stresses from work is critical. Having two of them sequentially provides even more benefit because for one of them you don't have to plan to get up the next day and go back to work.
The answer is obviously not 7 day work weeks, and obviously not 0 day work weeks, but is 5 on 2 off optimal, or is there room for improvement?
I have family who switched from 5 8 hour days (plus substantial overtime and on-call) to 3 12 hour days, and the difference in their health, mood, and energy levels is enormous. They get more done better at work than they did before, their job satisfaction is way up, and even their personal finances are night and day compared to before, even though they're making the same exact yearly dollar amount.
> 7-day workweek but a shorter workday to keep the # of hours constant
Why is 40 hours a sacred value? Why is it important that employees perform exactly 40 hours of labor per 168 hour period?
> eliminating the loss of productivity between weekends.
As a software developer with pretty free reign to do whatever helps the business the most, I would bet a paycheck that weekends create more productivity than they "lose". Over the weekend I've come up with ideas that saved hundreds of thousands of dollars a year. If I didn't have time to myself to relax and gather my thoughts, if every day was sitting there churning out code, I can't imagine I'd have even considered those things.
Thinking about it I bet I could even go over my past work and compare average productivity to weeks where I was just on-call over the weekend. I imagine there'd be a measurable drop after on-call weekends, even if I never was actually called.