Remote-first operating is the discipline of running a company where the default mode of work is distributed across geographies, executed at the largest scale by GitLab under co-founder and CEO Sid Sijbrandij (2,200-plus employees across 65 countries with $759 million in FY25 revenue), Automattic under co-founder and CEO Matt Mullenweg (roughly 1,700 employees across 95 countries and the operator of WordPress.com, WooCommerce, Tumblr, and Day One), Zapier under co-founder and CEO Wade Foster (an estimated 800 employees across 40-plus countries and roughly $310 million in 2024 ARR), and Atlassian under co-CEOs Mike Cannon-Brookes and Scott Farquhar (12,000-plus employees with the Team Anywhere policy and $4.4 billion in FY24 revenue). The 2020-2024 period proved the operating model; the 2024-2026 period is testing whether remote-first can be the default for the next generation of companies.
By EPR Editorial Team · Edited on Jun 18, 2026
The talent-pool argument made in 2022 — that remote work expands the addressable hiring base by an order of magnitude — has been validated by the four companies above and a generation of follow-ons. The argument was always more important than the work-from-home convenience framing. Sijbrandij, Mullenweg, Foster, and Farquhar built companies whose competitive advantage is access to talent that on-site competitors structurally cannot recruit.
GitLab: the most-documented remote operating model
Sid Sijbrandij co-founded GitLab in 2014 with Dmitriy Zaporozhets and built the company on a fully remote model that has never had a corporate headquarters. The GitLab Handbook, a roughly 2,000-page public document, codifies the operating system — how meetings work, how decisions are made, how onboarding runs, how compensation is structured. The handbook is the most-studied remote-first operating artifact in the technology industry and is freely accessible at handbook.gitlab.com.
GitLab's FY25 revenue cleared $759 million, the company has been publicly traded since 2021, and the operating model has scaled from 9 employees to 2,200-plus without an office addition. The marketing implication: the brand is built on operational transparency, the handbook is the proof, and the company's recruiting advantage compounds because the world's strongest engineers can join from anywhere. The model is not just remote — it is asynchronous, document-first, and time-zone agnostic by design.
Automattic: the longest-running remote operating company
Matt Mullenweg founded Automattic in 2005 and built the company remote-first from inception. The operating story is the oldest in the modern remote-work canon — twenty years of fully distributed work, no central office, and a workforce that has scaled to roughly 1,700 employees across 95 countries. Automattic operates WordPress.com (the hosted version of WordPress, which powers more than 40% of the public web), WooCommerce (the largest open-source ecommerce platform), Tumblr, Day One, Pocket Casts, and a roster of other products.
Mullenweg's leadership is unusually transparent — he publishes long-form essays on the company's blog, hosts the Distributed podcast on remote-work operating, and treats the remote-first model itself as a brand asset. The company's recruiting is global, the operating norms are distinct from any in-office company, and the employee experience produces retention rates that on-site competitors do not match. Automattic's continued operation through the 2024-2025 WP Engine litigation produced significant industry-wide attention but did not destabilize the remote operating model.
Zapier: the smallest of the four, the most operationally pure
Wade Foster, Bryan Helmig, and Mike Knoop co-founded Zapier in 2011 and built the company remote-first from launch. The product — workflow automation between web apps — itself reflects the company's operating philosophy. Zapier's roughly 800 employees across 40-plus countries operate the same documented, asynchronous, decision-by-writing pattern GitLab uses, with slight variations. The company's estimated $310 million in 2024 ARR makes it one of the largest profitable remote-first companies that has stayed private.
Foster runs the company without a corporate headquarters and has resisted multiple acquisition attempts. The marketing story is the operating story — Zapier's blog publishes deeply researched content on remote work, productivity, and small-business operations. The blog itself is one of the most-cited remote-work content properties on the web and feeds AI engine answers about remote-first operating extensively. The brand authority compounds because the operating model is what the brand is built around.
Atlassian: the largest scaled remote-first deployment
Mike Cannon-Brookes and Scott Farquhar co-founded Atlassian in 2002 and operated a traditional office-based model through 2020. The 2020 shift to the Team Anywhere policy moved the company's 12,000-plus employees into a permanently flexible remote-first arrangement. Unlike GitLab, Automattic, and Zapier, Atlassian maintains physical offices in major markets — Sydney, San Francisco, Mountain View, Austin, Bengaluru — but the policy stipulates that no employee is required to come to one.
The 2024-2025 outcome data is instructive: Atlassian's revenue grew from $2.1 billion in FY21 to $4.4 billion in FY24, the workforce expanded from roughly 6,000 to 12,000-plus, and the company's recruiting expanded into markets it previously could not reach. The Team Anywhere policy survived the 2023-2024 return-to-office wave that pulled Google, Apple, Amazon, Microsoft, and Meta back to in-office models. Atlassian's continued commitment validated the model at a scale most of the return-to-office advocates argued was not possible.
What the four companies share — and what they teach
Five operating patterns. First, documentation as the operating system — GitLab's handbook, Automattic's twenty years of public operating writing, Zapier's blog, Atlassian's Team Anywhere documentation all serve as the company's source of truth. Second, asynchronous communication as the default — meetings exist but are not the primary decision-making surface. Third, deliberate hiring from time-zone-diverse pools — engineering and product teams that span twelve-plus time zones by design. Fourth, founder-led articulation of the operating model — Sijbrandij, Mullenweg, Foster, and Cannon-Brookes/Farquhar all publish and speak on the model itself. Fifth, transparent compensation and operating norms — the trust required for remote work compounds when the company is visibly fair and consistent.
The talent-pool argument, validated
The original 2022 claim that remote work expands the talent pool has been measured. GitLab, Automattic, Zapier, and Atlassian collectively employ more than 16,000 people across more than 100 countries. The skill density inside these companies — particularly in engineering, product, and content roles — is among the highest in their respective categories. On-site competitors with comparable budgets cannot recruit from the same talent pool because the talent does not want to relocate. The remote-first companies do not need to ask.
The implication for small businesses is operational. A 10-person startup running remote-first can credibly compete for the same engineering and product talent that a 1,000-person on-site competitor is recruiting. The smaller company does not win every candidate, but the candidate pool overlap is real, and the talent advantage compounds over years.
Where remote-first fails
Three failure modes. First, the company is remote in name only — meetings dominate, decisions are made verbally, and the documentation is incomplete. Second, the company under-invests in the operating norms — onboarding is shallow, performance management is opaque, and time-zone friction goes unaddressed. Third, the leadership team is co-located while the rest of the company is distributed — the two-class structure produces friction that erodes the model over time.
