One of the questions I get most since joining Automattic is about what it's like to go from working with a company where we were mostly collaborating in one office space in Indiana, to working with a fully distributed company, where everyone works from home, coffee shops, co-working spaces or similar spots around the world.
(The other main question I get is about what it's like to go from being "the boss" to working for someone else - a post for another time.)
The short answer: I'm appreciating and enjoying it, and I think it is the future of many kinds of work.
The longer answer follows.
The question of a distributed versus in-person setup for a team or company is discussed often in tech circles, perhaps even more so right now as tech companies face hiring challenges and consider related immigration policy issues. I feel like I've experienced both sides in some form now, having built and managed a web development firm for close to 17 years with a strong focus on working together in the same physical space (experimenting with remote workers along the way) and now having worked almost 8 months in a fully distributed configuration.
Going in to my new job, I was initially skeptical that I would find a distributed setup to be better than what I'd experienced working with my team in an office together. I thought it would be exciting, interesting and different, and despite all of the enthusiasm I'd built up for it (especially after reading The Year Without Pants: WordPress.com and the Future of Work) at some gut level I still suspected it would be a kind of nominally acceptable, second-best alternative to working with people in person.
And indeed there were plenty of moments initially where I had the thought, this is crazy. "This would go so much faster if we were in the same space together." "I can't believe we're typing this conversation when a phone call would be so much more efficient." And so on.
(My colleague Steph Yiu just posted a great summary of the kinds of questions my team still faces about how to communicate best in a distributed setup, and you can imagine how overwhelming this configuration could be to someone coming into it new.)
But over time, I started to see the power of it.
I started to see how it was optimized to let people with incredible talent and self-discipline do what they do best.
I started to see how it eliminated many of the hiring frustrations and barriers I had faced as someone trying to grow a technology service business.
I started to see how it helped us move past some of the outdated limitations of the U.S.-centric work culture and the Monday-Friday 9 to 5 Eastern workday.
I started to see how the documentation processes that inherently happen in a well-done distributed workflow help to build and preserve institutional knowledge.
To be clear:
In an ideal world, I think having your entire team in the same physical space is still the best way to go.
Humans are still social creatures that have evolved to communicate in ways that benefit significantly from being in person. Our facial expressions, tone and body language make up a big part of what we're saying, in ways still not fully captured by any video conferencing technology. Our ambient awareness of each others` locations, moods, activities and work habits are important pieces of a team's culture, still not fully captured by any online collaboration tools. And there's just a richness to being in the same physical space that cannot currently be replicated across the miles.
But the "ideal world" part of the above is important.
In an ideal world, you'd have a physical work space so flexible that it would meet the changing needs and size of your team as it evolves and grows, without investing lots of capital in renovations and expansions.
In an ideal world, the place where your office is located would offer the perfect surrounding landscape to satisfy all types of workers and the personal interests of them and their families.
In an ideal world, the city and state you choose as your organization's home would have the perfect mix of economic, educational, political and social resources and norms in place to attract and retain just the right kind of potential employees you want to have.
In an ideal world, your organization's attractiveness as a place to work and the benefits you offer would be so self-evident that all other considerations about the mechanics of employment there would always be secondary.
In reality, I don't know of a place on Earth that can meet all of these criteria in perpetuity.
Most places of work and their employees have accepted this, willing to make trade-offs to get most of the way there. We commute a little longer than we'd like to work somewhere that recognizes our talents and passions. We take a job in a city that isn't quite paradise because our spouse/partner can also find a job there. We work in a space that isn't quite conducive to full enjoyment of our days because the pay is good and the work is interesting. Employers hire people who aren't quite the best person for the job because they're the person who's available in that city at that moment to do the work.
This is where the philosophy of building distributed teams comes in and tries to make those compromises unnecessary.
"What if you could hire the best possible person for the job, let them live in the best possible location for them and their family/community, let them work in the best possible space for their personal productivity, compensate them in the best possible way to reward their talent, and retain the flexibility to quickly adapt your entire infrastructure to changes in the business landscape?"
There are many kinds of organizations that can't be distributed. But for any that can even consider it, it's a pretty appealing way to go. Having been immersed in one for a while now, I can safely say that if I were to create a knowledge-based company again I would almost certainly do it distributed.
Based on my experience so far, there are a few caveats I'd share with anyone considering building a distributed team:
You have to hire people who can work at 100% on their own. Distributed teams aren't a good fit for every kind of person; that doesn't mean they're good or bad, it's just about personality and what works. Some people may have a really hard time sitting in a room alone for hours on end (though you can mitigate that somewhat by encouraging the use of co-working spaces and coffeeshops). Some people want or need prescriptive and regular supervision of their daily work. Some people will find they're not passionate about the work and, with no one to hold them accountable in the moment, will minimize the personal energy they put in to it. And some people who might already have a hard time asking for help or admitting they don't know something would probably find the workflow of a distributed team to be brutal.
To make distributed teams work, you need people who take full ownership of their roles, who take initiative in figuring out what's expected of them and delivering it, and who actively seek out the right balance of productive work time and rejuvenating personal/social time without having the rhythms of office life to go by.
You have to quickly remove all administrative, technical and logistical barriers to distributed work. You can't say that you're going to build a distributed team and then ask those workers to jump over a bunch of hurdles or inefficiencies that hail from the working-in-person world. All of your legal and financial paperwork needs to be done electronically. You need to have great communication tools set up from the start. Your VPN and related security systems need to be easily configured and compatible with the tools your team wants to use to do their job most efficiently. You need to invest in equipment and environmental considerations that help make distributed workers effective.
Every part of your operations needs to be tailored to support the distributed work life. If you do it half-way, I don't think it will work.
Having some remote workers is harder than being fully local or fully distributed. Some organizations dabble with the benefits of having parts of their teams be remote while others work in central locations together. I think this is actually harder than being fully co-located or fully distributed.
For one, it means you have to keep investing in dual approaches to your infrastructure, policies, and guidelines. Every time you consider an organizational change, you'll have to think through the implications for the in-person worker experience AND the remote worker experience. You'll face concerns about one type of worker getting more perks or benefits from one kind of approach than others do.
More importantly, though, this dual approach is probably a recipe for disaster when it comes to building shared vision and common culture in an organization. If there are team members who have a daily experience of being in the same space together and sharing all of the quirks and benefits of that, remote workers will almost always feel excluded in some way, culturally, logistically or both. When only part of the team is forced to consider the implications of having a distributed group, an unfair burden falls to the remote worker to keep their needs in front of everyone else. At best it adds a weird kind of tension to team relationships, and without incredible discipline and initiative, it probably won't work in the long run.
These notes don't even begin to explore all of the details and considerations of building a distributed workforce, and many of those have been covered in various books, articles and posts elsewhere. I'm sure my own thinking on it will continue to evolve, too.
I enjoy writing about this kind of stuff, so if there's some aspect of my take on distributed work you'd like me to go into more detail on, just let me know. And if you have experiences with distributed work, I'd appreciate getting your take on what works and what doesn't.
I’m a journalist, publisher, software developer and entrepreneur with experience as a founder and organizational leader. Work with me or learn more about me.
Nice post! How would you square up the end of your post with the fact that we have Hawthorne? (I have my own thoughts on this, but interested to hear yours!)
It's a good question! I don't know the full history of how Hawthorne came to be or all that it does, so some of this may be inaccurate, but: I describe it to people as a company lounge that we use for a variety of purposes and events, but that isn't necessarily critical to the company's operations.
I think any organization that saw it was regularly hosting people or conducting events in a given location would want to consider reducing expenses by investing in a dedicated space, so it makes sense that Automattic did. I suppose the natural extension of that logic is that as an organization gets big enough and has enough of its people in the same place, an office building or even full-blown campus just makes sense, too (e.g. Google or Apple).
I agreed with everything in your piece up until "More importantly, though, this dual approach is probably a recipe for disaster when it comes to building shared vision and common culture in an organization". I work in a fully distributed company (there's only 4 of us though - Belgium, Scotland, North England, South England). However most existing companies will be growing out of a fully co-located environment. I don't think any company can fully switch over to be distributed in one step - so they have to go via the dual approach. I have worked with two clients that I've worked with for more than 4 years, both relatively small companies (<30 people) who have about 10% of staff working remotely. Its not optimal and tools like Slack don't get utilised, but its certainly not a disaster.
Ian - a fair point, and "disaster" was unnecessary hyperbole. I'm sure an organization that takes a thoughtful and intentional approach to a mixed setup can do just fine, especially if they're transitioning from one clear structure to another. I guess I've seen too many cases where there wasn't much thought given to it in the eager rush to take advantage of remote work benefits, and the experiment ends up failing, sometimes at the great expense of the remote workers themselves.
I'll look for more transition/dual-setup success stories to balance that out.
I work for a company that has been taking a dual approach for more than a year now, and I feel we are being very successful with it. The difference for us is that it isn't one or two people from various departments - the entire engineering and design team is remote, while product, sales, and marketing are local.
And while I would say that there is some separation between the locals and the remotes, I would also say that the separation isn't that much different than what I have experienced between Sales and Engineering teams when both were in the same building.
However, having the entire Engineering team be remote helps form as strong a bond within the department as would happen if we were all local. And to keep us connected to the company, the remote team comes in quarterly, and has monthly town halls with the CEO.
So I feel that a dual culture can be successful if done right.