Remote vs Co‑located Work: Multiple Team Distribution Models and Their Trade‑offs
The article argues that remote work and co‑located work are not a simple binary choice, outlines several team distribution models—single‑site, multi‑site, satellite workers, remote‑first—and discusses their productivity, communication, and agile implications, concluding with practical recommendations for managing distributed software teams.
Author: Martin Fowler Translator: Dong Ge Original: https://martinfowler.com/articles/remote-or-co-located.html
Remote work and co‑located work are not a simple dichotomy; teams can adopt many distribution patterns, each with its own trade‑offs and suitable techniques.
One of the most profound impacts of the information age is the ability to accomplish tasks without regard to geography, a fact especially evident to software developers who are at the forefront of digital transformation.
Despite the hype around bringing all staff back to a single site—Yahoo, Netflix, Google—many successful internet companies such as Etsy, Basecamp, and GitHub have operated with distributed teams for years, proving that remote work can be the future for many.
There is no single decisive factor that proves remote work is better or worse; instead, the author shares observations gathered from many teams.
1. Many Forms of Remote Work
• Single‑centered team : Everyone works in the same physical location, enabling instant, face‑to‑face collaboration.
• Multi‑site team : Two or more locations form a larger team, often with sub‑team boundaries (e.g., Melbourne‑Xian).
• Satellite workers : Most members are co‑located, while a few work from home or other offices.
• Remote‑first team : Everyone works remotely (usually from home), with all communication online; most open‑source projects follow this model.
• Degrees of remoteness : Even splitting a team across two floors can diminish the feeling of co‑location; greater distance and time‑zone differences increase the sense of remoteness.
2. Productivity in Co‑located Teams
The author cannot provide quantitative evidence that remote work reduces productivity; anecdotal claims that co‑located teams are more efficient are common but not definitive.
Many factors influence team performance, and comparing a single‑site team to a multi‑site one without controlling for other variables is unreliable. Observations suggest that communication ease in a single site often leads to higher productivity, but this is not the only consideration.
3. Remote Teams Can Be More Efficient
While single‑site teams may be more productive, they limit the talent pool to those willing to relocate. Remote models broaden the talent pool, potentially yielding higher overall effectiveness despite a possible drop in per‑person productivity.
Remote work also offers personal benefits—flexible schedules, reduced commute, better work‑life balance—and can improve diversity, especially for caregivers.
Off‑shoring expands talent access, as illustrated by ThoughtWorks leveraging talent across continents.
4. Communication Patterns
Effective communication is the core of software development. Remote work changes communication channels, often reducing the richness of face‑to‑face interaction.
Multi‑site teams may develop "us vs. them" attitudes; regular visits and ambassador programs can mitigate this.
Remote‑first teams should conduct all communication online but still schedule periodic face‑to‑face gatherings to resolve complex issues and strengthen relationships.
Work should be organized around autonomous, full‑stack components rather than layered divisions, respecting Conway's Law.
Satellite workers are challenging to integrate; regular on‑site visits and clear autonomy are essential.
Mentoring junior staff remotely is difficult; ensure experienced mentors are present at each site and avoid assigning junior developers as satellite workers.
5. Remote Work and Agile
Some argue Agile and remote work are incompatible, but Agile values individuals and interactions over processes and tools, encouraging the recruitment of the best talent regardless of location.
While face‑to‑face communication is valuable, it does not negate the importance of remote collaboration.
Conclusion
There is insufficient evidence to draw a definitive conclusion about the effectiveness of remote work. Key take‑aways:
Teams have many distribution models beyond a simple remote vs. co‑located binary.
Co‑located teams often communicate more richly and can be more productive, but some individuals thrive in remote‑first settings.
Adopting a distributed model can broaden the talent pool and improve overall team effectiveness.
When using remote work, pay close attention to communication patterns and invest in tools, travel, and relationship‑building activities.
Organizations that master remote work will gain a significant and growing competitive advantage.
DevOps
Share premium content and events on trends, applications, and practices in development efficiency, AI and related technologies. The IDCF International DevOps Coach Federation trains end‑to‑end development‑efficiency talent, linking high‑performance organizations and individuals to achieve excellence.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.