The Costs and Benefits of Distributed Teams
The Costs and Benefits of Distributed Teams
Normally, when people consider the pros and cons of distributed teams, they are evaluating whether supporting remote work is worth it. That’s not a question for me. I have worked on distributed teams for over 15 years — most of my 18+ years as a professional in my field. Of those 15, I have been working remotely as an engineer and leader for over 10. I’m a fan.
My interest is instead in building a perspective for evaluating risks, problems, choices and processes to support distributed teams. As a leader in a distributed organization, I want to reduce the costs and amplify the benefits of remote workers. So what are they?
Before we dig in, let me define some terms.
I’ll use “fully distributed” to describe teams and organizations that are 100% remote. Examples include GitLab (more than 140 people in 2016), Automattic (>580 people in 2017), Zapier (>80 people in 2017), Invision (>150 people in 2015), Articulate (>150 people in 2017), Aha! (>50 people in 2015), Doist, Treehouse, Balsamiq, and more.
“Co-located” describes a standard office environment. There are thousands of them, but IBM and Yahoo! make especially pertinent examples, since they rolled back their support for distributed teams relatively recently.
Finally, we have “partially distributed” teams, which mix remote and co-located workers. My company, manifold.co, fits this description, as do many others, such as GitHub (65% remote in 2015), Trello (65% remote in 2017), Canonical (75% remote in 2015), Red Hat (25% remote in 2017), Basecamp, Elastic, Dell divisions, Microsoft divisions, Salesforce divisions, and many more.
With those terms as a guide, we’re ready to begin.
We’ll start with the costs. Many of these costs have a variety of tried-and-true mitigation strategies — technical, process-oriented and cultural — but that’s out of scope for this post.
The central difficulty with remote work is that synchronous communication is harder for distributed workers than for co-located workers, and this makes personal connections harder. In my experience, email, group chat, phone calls, and video calls don’t seem to grow relationships as well as communicating to someone in the flesh. That reduces trust and empathy, and makes mentoring and onboarding harder. There’s also no physical water cooler or lunch spot around which to share ideas spontaneously. Virtual whiteboards aren’t as easy to use as real ones, which limits brainstorming and collaboration. And awkward audio lag can also reduce group participation.
Partially distributed teams have it even worse, because a co-located group can easily and unintentionally hog the conversation, while remote workers try to fit a word in edgewise, with audio lag, echoes, and echo-cancellation hindering their participation. And visually, conference room calls with multiple people in one camera make it difficult for speakers to quickly get a feel for reactions in the rest of the room.
These challenges are significant already, but the difficulty with synchronous communication and personal connections has some really important corollaries. People without sufficient social face-to-face support in their personal life (family, friends, social activities) can become lonely and isolated. Managers and coaches have somewhat less opportunity for visibility into what’s going on. People without an internal drive to deliver probably won’t succeed.
Finally, again for partially distributed teams, if a large percentage of a team, or even just most of the technical or managerial leadership, is co-located, the remote workers are at a serious disadvantage. They are very likely to be isolated, and miss out on important conversations, or even just the rumors and the collective emotional “feel” in the office. Moreover, they will have a harder time getting recognized, which can lead to difficulties in promotions and growth opportunities.
There are a couple of additional complications to supporting distributed teams. They aren’t clearly costs, but they are worth calling out.
First, fair, comparative salaries are harder to determine when you have remote workers, and employees may feel unfairly treated once their pay rates are revealed. Should you use a service like PayScale to account for different pay rates in different locations? Should you also factor in cost of living? How should you handle very desirable candidates with a lot of remote experience who are accustomed to higher-market salaries than where they actually live? What do you do when someone decides to relocate from high markets to low markets, or vice versa? How should you balance differing taxes, vacation and health care around the globe? GitLab has done a real service to the community of remote companies by transparently documenting their policies and a salary calculator and I like what they have done. I personally still find it unsatisfactory because of some of the questions I raised above, but haven’t yet identified concrete changes I might make to address them.
Second, sometimes people without distributed experience think they will like working remotely more than they actually do. When they try it, it leads to unhappiness, underperformance and disappointment. Some people simply prefer an office.
With all of those costs and complications, why would a company bother supporting distributed teams?
From my perspective, and as described in many of the remote.co company questionnaires, a key benefit is hiring. A remote position can open the door both to people who live in different parts of the world, and to people who need to work from home for any number of reasons. Organizations can find a significantly bigger pool of highly qualified candidates if they are willing to support remote workers. This is true if the company is based in any of the largest cities in the world. It’s even more true if the company isn’t.
Another win is that a good remote culture is a big perk and retention benefit for many employees. Successful remote workers typically appreciate the lack of a commute, and the flexibility and control of working hours, workspace, and residence. Eliminating a commute or allowing someone to live in their desired location can dramatically change a life.
Finally, a distributed culture also helps diversity in any many ways. Greater international diversity can guide the company to better understanding of international markets in their product and marketing. Greater timezone diversity can reduce operational and support burdens by encouraging hand-offs, and not having to wake people up as much, or keep people up late. And greater diversity of backgrounds can be a great context for practicing empathy and leveraging new problem-solving perspectives.
I have come to be somewhat skeptical of a couple of other possible benefits.
First, a potential benefit is that remote workers can sometimes cost less. If you compare salaries in the midwest of the US against a San Francisco salary, for instance, the difference can be substantial. Also, organizations don’t pay for regular office space for remote workers.
That said, supporting distributed work often has hidden costs. You might provide an initial stipend to outfit a home office, and also regularly reimburse home internet or even co-working spaces. You might be investing more heavily in collaboration cloud services and home-grown communication tools than you would otherwise. If you don’t provide the right support or hire people who can work well in a distributed team, reduced communication and connection may decelerate your team. You might pay non-trivial transportation and hotel fees to get your teams together every few months. Each country you operate in will have at least some legal and operational overhead. And if you have an office, you might even pay for your office to have at least a few floating desks so remote workers have a place to work when they come into town.
Second, another potential benefit is that eliminating commuting can be a potential gain for the environment. It’s true; but on the other hand, if someone flies long distances even a few times in a year for trips that she would not have made if she were remote, the high environmental cost of aviation can offset or even overcome that potential benefit.
Reduce the Costs and Amplify the Benefits
To my mind, these costs and benefits are significant. A fully distributed team must examine and address these costs and benefits as a matter of survival! But what about partially distributed teams? Do they need to concern themselves with these problems as carefully?
I think the answer is a strong “yes.” Partially distributed teams exist along a spectrum between what I call “remote reliant” (sometimes called “remote first”) and “remote friendly.” On one side, remote reliant organizations deeply rely on and support remote workers using practices shared with fully-distributed companies. On the other, remote friendly teams operate more like standard co-located teams, with some extra remote people. Most partially distributed companies will not fully match one side or the other, but it’s worth considering where you reside on the spectrum. Having worked on several points along this continuum, I find the remote reliant side to be far more successful. Otherwise the remote workers will run a high risk of being isolated, unproductive, and unhappy — which is bad for everyone.
If you do decide to make the choice to work remotely, or to support remote work, I encourage you to ground your plans in some of the considerations discussed above. Reduce the costs, amplify the benefits, and build a company culture that can flourish around the world!