top | item 32047778

Microsoft researched what made employees happy

201 points| alok-g | 3 years ago |zdnet.com | reply

171 comments

order
[+] haswell|3 years ago|reply
I think there is a major framing issue here.

Speaking for myself, and the many colleagues I’ve spoken to about this, collaboration isn’t the issue.

“Collaboration” is the issue.

I’ve had some of my most meaningful (and happiness-inducing) experiences at work when collaborating deeply with colleagues.

I’ve had some of the most frustrating experiences at work when “collaborating”.

The distinction simply being: “Collaboration” in the form of “interacting in realtime” or “existing in the same space” can also be 99% distraction.

“Real” collaboration - the kind where the collective consciousness of the group results in outcomes that could not have been achieved solo - is where the magic lies.

I’m absolutely happier with more autonomy and less distraction. But that is not the antithesis of collaboration.

A pendulum that swings too far in the other direction will leave folks feeling isolated, and will reduce the potential of a group. Finding the right balance and distinguishing between real collaborative interaction vs. “acting out” something that looks like collaboration on the surface is critical.

[+] matwood|3 years ago|reply
They probably could have just shortened it to autonomy. Collaboration is great when it’s not forced - hence autonomy.

I also like to call autonomy to being treated like an adult. So much of what companies do is infantilizing now.

[+] closeparen|3 years ago|reply
Collaboration implies underlying work. Something the collaborators are doing together in addition to talking. The vast majority of the talking in my corporate environment either has no such referent, or is a one-sided accounting of the work you are doing up to someone whose job it is to allocate your time or supervise that work, not to personally advance it with you.

A good litmus test is what someone's calendar looks like. People who are in wall to wall meetings every day (managers, Staff+ engineers, TPMs, etc) can never, properly speaking, collaborate, since there is no time where they could be doing the work being collaborated on. They have coordination and supervisory functions that may be valuable, even indispensable, but what they do is different from collaboration.

[+] mrexroad|3 years ago|reply
Too many orgs conflate communication and collaboration. The heavy handed push for “Collaboration” always seems to stem from folks with anxiety around feeling left out, and leadership who is equally disconnected. E.g. “I’m hearing you’re not being very collaborative, spend more time bringing others along!” When really the problem is lack of communication and transparency creates fear/uncertainty/doubt. With that said, communication and coordination are the foundations for collaboration. Some orgs address those well, most don’t; too often it falls on the right interpersonal dynamics to create the environment for meaningful collaboration where 1+1=3.
[+] xboxnolifes|3 years ago|reply
"Collaboration" is the kind where someone read an article about a team building a successful project with lots of collaboration once, so they mandate pair programming on every ticket and mob testing for every PR.
[+] jokethrowaway|3 years ago|reply
Collaboration has its place for when you need it. Sure, you can get great results despite paying the mental cost of collaborating (which is even higher if it's happening through a computer screen).

The problem is that we see the great results of occasional collaboration and we want more of that. Creating more artificial collaboration when you're capable of doing the task by yourself will just be expensive (both in terms of salary and mental health): it's not the answer.

[+] tikhonj|3 years ago|reply
Exactly the same for me: some of my best work and most enjoyable time has been collaborating closely with people. And yet, some of my worst time has been in settings that can also be described as collaborative!

What made a difference? I see it as having two types of "collaboration": working together vs working on the same thing in parallel.

The problem is that the formal processes managers use to ostensibly encourage collaboration—and, of course, to track and direct individual work—push hard towards the latter model. Everybody working on their own tasks makes the work so much more legible! Having some other person or some process determine who works on what, when and then implicitly (or explicitly!) judging people based on how quickly they complete "their" tasks is not only awful for autonomy and job satisfaction but also actively makes it harder to collaborate meaningfully. I've worked on teams where people were afraid to spend too much time working together or helping each other because they didn't want to "waste" the other person's time—after all, you're clearly wasting time if you aren't consistently producing finished tickets!—which is absolutely toxic for creativity, culture and collaboration, but also a natural response to the incentives and structure imposed by the process.

I'm gradually starting to look for my next job and finding a culture where I can actually collaborate is one of my main goals, but it seems hard to evaluate from the outside. It's not clear how much an individual manager can do to foster the right sort of environment—a lot depends on the culture and structure at higher levels on the organization, as well as cross-team interactions—and it's the sort of thing where everyone is going to say their team is collaborative (and even believe it), and I don't see how a team could show rather than just say that.

[+] more_corn|3 years ago|reply
Me in mind that at Microsoft and every large company. A lot of time is wasted in meetings. What you’re seeing is that the happiest workers AT Microsoft are the ones with more time for deep work and the ones with high autonomy. That lesson is probably generalizable. I don’t think anyone is suggesting an end to collaborative work.
[+] agumonkey|3 years ago|reply
a good collaboration aims at making parties very independent at some points. because you all shared the same outlook and plan and can now go on your own way until the next mental rendez-vous.
[+] clumsysmurf|3 years ago|reply
> Does it, in fact, intimate that collaboration may have become a buzzword for a collective that is more a bureaucracy than a truly productive organism?

I recently worked at a company that hired Agile consultants to install SAFe. We had PI planning each quarter, which ran for 3 full days. At its conclusion, the consultants, leaders, and PMO always called it "successful collaboration." I guess they were having lots of fun ... but the engineers were not. Most of it was us guessing estimates for things 6 weeks out, and management turning that into commitments. Needless to say, this whole circus became negative experience for almost all the producers.

[+] bdavisx|3 years ago|reply
LOL, safe uses an "agile release train", because everyone knows how agile trains are. They can stop on a dime and move in any direction, the very definition of agility.
[+] wsostt|3 years ago|reply
How do you propose stopping estimates from becoming commitments? Genuinely asking because I struggle to not do this as a manager.
[+] tpmx|3 years ago|reply
My experience: Collaborating closely with smart people makes you happy. It can be so exhilarating - in a sustained way for a long time, if you're lucky and the universe aligns correctly.

And yes, frequent interruptions are okay in this context. You get into a rhythm.

[+] drewcoo|3 years ago|reply
Why do they want happy workers and how do they measure happiness?

When exactly did Microsoft abandon all the lessons of Peopleware[1]?

And what about flow[2]? I recall people blocking out calendar time and putting do not disturb signs on their office doors[3] when they needed time to get into flow. Good managers helped make that happen. Does that not happen there anymore?

[1] https://www.goodreads.com/book/show/67825.Peopleware

[2] https://www.quora.com/What-is-a-flow-state-of-mind-How-succe...

[3] We had actual offices, doubled up and individual. They had doors. It took almost 2 decades to finally work remotely so that I could have an office with a door again, this time in my home. Bliss!

[+] bawolff|3 years ago|reply
Is this really all that surprising? In school everyone always complains about group work sucking.
[+] ergocoder|3 years ago|reply
Group work has all the bad attributes:

- can't be fired

- reward is low

- members often have the exact skillset.

- nobody has superior authority. How does one even make a decision where people disagree?

Teachers have no experience working in the real world and will often say "you will need to learn to work with bad people".

Meanwhile every successful person will tell you to fire bad people as soon as possible.

If you assign a bunch of low performers to Steve Jobs and tell him he can't fire them and have to give them work, even Steve Jobs would probably fail.

[+] haswell|3 years ago|reply
I think this is a good anecdote that reiterates the kind of “collaboration” the article is talking about, but is not the “real” collaboration people strive for.

The reason group projects in school felt so terrible was because the end goal was collaboration, not some higher outcome. Collaboration can unlock outcomes that can’t be achieved as an individual, but collaboration can’t be the desired end state.

[+] yodsanklai|3 years ago|reply
> In school everyone always complains about group work sucking

I supervised software engineering projects at university. It works well when students in a group have similar motivation and abilities. This is the case when they are free to choose who they collaborate with. Otherwise, students are usually frustrated.

In a company, teams can be much more heterogeneous (cultural background, skills, age, social status, experience...), and the pressure is higher: goals can be loosely defined, evaluation is less fair, stakes are higher for everyone, managers can be less benevolent than teachers...

[+] closeparen|3 years ago|reply
It is perhaps surprising to a management class that believes collaboration is the ultimate good in all circumstances and you can never have enough of it.
[+] pjmorris|3 years ago|reply
FTA:

"By combining sentiment data with de-identified calendar and email metadata, we found that those with the best of both worlds had five fewer hours in their workweek span, five fewer collaboration hours, three more focus hours, and 17 fewer employees in their internal network size."

[+] markmark|3 years ago|reply
So if they are looking at calendar data does "five fewer collaboration hours" only mean five fewer hours in meetings?

Some of the funnest and most productive work I've done was a project with three others where we were all in a single office. We were completely ad-hoc with working alone, pair programming, having quick design discussions, whatever was needed at the time. There was an enormous amount of collaboration (and an enormous amount of deep, solo work), but there were no meetings in my calendar.

Meanwhile in pretty much any job I have had my happiness has gone down the more full my official calendar becomes.

[+] dredmorbius|3 years ago|reply
It would be awfully useful to know what the baseline from which the 5 & 17 fewer reductions are based on.
[+] lucb1e|3 years ago|reply
TL;DR^, came here to post the same quote (they are citing the original authors here). The rest of the article is filler and speculation based on this statement.
[+] jokethrowaway|3 years ago|reply
I dumped my last job exactly because collaboration became the mantra.

I tried to effect change but leadership didn't listen and my team wanted to apply the collaboration koolaid and be good team members.

Glad there's research I can point to.

[+] just_boost_it|3 years ago|reply
The structure of the experiment seems to have been just a correlation between collaborative hours and some measure of happiness. I wonder if the causation doesn't run the other way to what they propose though. When I'm happy in work, I don't really feel the need to have to work to grow my network outside my team. I have worked a job though where things just weren't clicking, so I did put more effort into trying to make them click by meeting people. That would have included getting involved in more projects, and doing a bit more collaborative work.
[+] drewcoo|3 years ago|reply
This.

Things aren't going so well at work? Work harder. Work more. Then burn out and quit, of course.

[+] sedatk|3 years ago|reply
I reckon that it's not the collaboration itself makes you unhappy, it's the contemporary methods used such as constant interruptions over a multitude of communication tools, frequent brain-melting meetings with no obvious purpose, not having a solid leadership leaving the team members to their own devices.
[+] thenerdhead|3 years ago|reply
By reading the article, the word "collaborating" somewhat felt like they were referring to meetings based on the word used around the context of people's calendars and availability that week.

Having less meetings and more individual contributor time will make anyone happy. You can at least do the job with that time and not stress that you will get it done that week with whatever artificial constraint that week has created based on the "collaboration" you were involved in.

People like determinism, no surprises here.

[+] cosmiccatnap|3 years ago|reply
I think it should be healthy to be allowed to simply do your job if you are an engineer. If you are a good one you'll know when you need help and when you need to grind.

Instead you are constantly in a state of being micro managed by people who think that if they have lots of meetings and gather lots of metrics that will increase productivity but it's a silly idea when you think about it

The only thing that gets work done is well...doing work and any amount of time you take away from that isn't productive 9/10 times.

[+] gerdesj|3 years ago|reply
What this article is dancing around is: Our employees are doing fuck all or just enough and we have managed to create an environment that actively encourages this behaviour.

You'd (not) be surprised how often this happens. $BIG_TECH hire people clever enough to look after themselves as well as their employer.

Now, what makes people happy? This doesn't:

https://hbr.org/2022/06/why-microsoft-measures-employee-thri...?

[+] underwater|3 years ago|reply
Some companies have a culture that favours meetings and cross-team builds over documentation and internal platforms. It might just be that someone doing more collaboration is working on more ill-defined problems, or are at a different part of the development cycle.
[+] spywaregorilla|3 years ago|reply
I've done some work adjacent to this. Not this project, not with microsoft. By internal network size they probably mean people you have a lot of traffic with in meetings and emails. Presuming they're competent they probably only look at interactions with small groups or 1:1. By collaboration they just mean meetings with coworkers.

Without having read in detail, its likely just drawing correlations between communication metrics and survey results.

Lots of companies are paralyzed big their meeting culture. It's not really a surprise to anyone. But the data can paint it a bit more clearly if people don't know quite how to articulate the problem. It's pretty much always the same in a large org.

[+] Groxx|3 years ago|reply
Is there a link to the study anywhere? I'm curious what kind of "internal network" survives removing 17 people while being more productive.

Certainly some do. But it means very very different things if their network was closer to 20 people or 200+.

[+] jpswade|3 years ago|reply
It says “five fewer collaboration hours”. It’s not clear how many hours of collaboration we’re talking.

It’d be interesting to see the actual figures before drawing a conclusion.

On one hand I can imagine if you’re spending all your time in meetings it would be frustrating, meanwhile I’ve what happens when teams don’t collaborate, it’s not a pleasant experience, it just causes tension to build up.

[+] outside1234|3 years ago|reply
They should research paying people at market next vs. not (current state) to see what the difference is :)
[+] Madmallard|3 years ago|reply
I don't know if I would trust a scummy anti-customer company to know what works best for people.