How Buffer Runs a 75-Person Remote Company With Transparent Salaries and No Office

Snapshot

Buffer is the social media management tool. Its December 2025 shareholder update reports 198,537 monthly active users and 69,760 paying customers. These are dated operating results, not today’s live metrics. (Source) The company reported 75 profit-share recipients in February 2026, supporting an approximate early-2026 team size of 75. (Source) Joel Gascoigne announced the decision to go fully remote in 2012; Buffer’s last physical office closed in late 2015. (Source)

Buffer’s standout contributions include financial transparency (public salaries since 2013) and operating documentation that explains its choices and revisions. (Source)

Their 2023 post “Buffer is Remote but not Async-First, Here’s Why” is one of the most candid pieces of remote-work operating documentation written by a practitioner. It is worth reading directly.

Core Philosophy

Buffer’s published communication philosophy emphasizes judgment about when to use synchronous or asynchronous communication, with flexibility in both directions. (Source)

Buffer’s current published values are Default to transparency, Improve consistently, Be a no-ego doer, Show gratitude, Choose optimism, and Reflect to unblock. (Source)

Inference: a public salary system makes it easier to inspect whether pay follows the company’s stated principles. It cannot by itself establish that every employee considers the outcome fair.

Communication Model

Buffer is explicitly not async-first. This is their own description, not a criticism. From their published documentation:

“Some remote companies are async-first. A few are even fully async with no live calls or in-person time at all. However, this isn’t what we’ve chosen for our culture.”

(Source)

Buffer’s published 2023 communication framework — three-way split by tool:

(Source)

Why they keep synchronous communication:

1. Speed for complex problems. Waiting for async replies during a crisis is genuinely worse than a 30-minute Zoom call.
2. Relationship building. Video calls humanize people in a way that written communication does not. Facial expressions, tone, and real-time response build trust that text cannot.
3. Things that only surface in live conversation. “In live conversations, sometimes topics come up naturally that wouldn’t have otherwise.”

(Source)

Slack usage has explicit norms — Buffer has published “10 Slack agreements” governing expected response times, use of notifications, and what belongs in Slack vs. elsewhere. Slack is not abandoned; it is governed. (Source)

“Give flexibility to get flexibility” is the summary principle. Team members are expected to occasionally stretch their hours to accommodate others’ time zones — and in return, they receive the same flexibility. It is not a one-way demand. (Source)

Planning and Cadence

Buffer’s public operating accounts describe several cadences and learning practices:

• Profit share tied to annual performance. In February 2026, Buffer reported its seventh profit share: $377,005 for 75 teammates, based on 15% of 2025 net profit. Allocation combines equal, salary-based and tenure-based portions. (Source)
• Transparent shareholder updates. Monthly public reports to shareholders covering revenue, customers, active users, and profitability. These are not marketing documents — they include negative months. (Source)
• Four-day schedule with a flexible fifth day. Buffer’s current careers page describes four workdays and a meeting-free fifth day available for overflow work or personal time. (Source)
• Company Build Week. The August 2026 AI build event is documented in the AI section. It is a one-week output count, not the planning cadence.
• Preserve context during cleanup. In March 2026, Buffer described tracing unused background jobs through Git history and old documentation, checking new logs for several days, removing jobs incrementally, and recording why they were removed. The team said it was adding dependency tracing during deprecation and responsibility documentation when people leave. These were reported actions and intended process changes, not a quantified maintenance result. (Source)

Decision-Making Model

Buffer’s decision-making model is not publicly documented in detail, but their transparency norms give it shape:

• Public salary system makes the calculation inspectable. Buffer’s 2024 explanation separates market data, internal role-and-level benchmarks, and individual salaries. It applies high/global location factors and retains a legacy salary-choice adjustment for some employees. (Source)
• Co-founder Joel Gascoigne (CEO) makes major decisions and publishes reasoning publicly via shareholder letters and blog posts.
• Ownership culture. Buffer’s current values emphasize initiative, adaptability, and considering the wider team. (Source)

Org Structure

• ~75 employees as of early 2026. (Source)
• Traditional structure with executives, managers, and ICs — not manager-less.
• Multiple continents. Teammates are listed by name and country on the public About page, which currently shows the team in countries including the USA, Canada, UK, mainland Europe, Australia, Nigeria, South Africa, and Singapore. (Source)
• Fully-remote decision in 2012; last physical office closed late 2015. The office was abandoned deliberately, not closed by circumstance. (Source)
• Buffer’s About page lists Joel Gascoigne as CEO and Hailley Griffis as Director of Communications & Content, accessed November 1,2026. This observation does not establish when her title changed. (Source)
• Hiring context before the first interview. Buffer sends a role-specific Q&A document when inviting candidates to interview. The practice began in engineering and had spread to most open roles by July 2026. (Source)

Tools and Stack

(2023 communication framework, February 2025 project-management update)

Rituals

In-person connection. Buffer’s 2023 communication account says the remote team also meets in person. It does not establish a current fixed annual retreat cadence. (Source)

Profit share. Annual distribution of 15% of net profit to all teammates. This is a ritual that reinforces the “we win together” operating principle. (Source)

Monthly public shareholder reports. These are as much an internal ritual as an external one — they force honest accounting of performance every month, publicly, which shapes internal discussions about what matters.

Transparent salary system. The published benchmarks and calculations make compensation an inspectable operating practice. (Source)

AI

The content function had earlier boundaries. In March 2024, Buffer’s content team documented AI use for ideation, outlines, proofreading, and repurposing, with human editing and accuracy review. It said it did not use AI for research or fully drafted articles, and described a marketing lunch-and-learn. This is a dated functional account, not verification that every boundary remained unchanged in 2026. (Source)

Every teammate can expense AI tools. On February 28, 2025, Buffer made all teammates eligible for $250 per year for AI tools of their own choosing, after the stipend began as an engineering discussion. The post encourages sharing experiments. It does not report uptake, savings, or whether that exact allowance was unchanged later. (Source)

Build Week 2026 put that access into a company-wide build. The August 28, 2026 recap describes the prior week: 72 people stopped regular work and each directed an AI-assisted project. Earlier Build Weeks used teams and pre-selected projects. By Friday the company reported 112 projects completed, 57 internal tools live from 41 people, 1,083 pull requests merged across 26 repositories, 15 changes shipped into the admin product, and six first-ever pull requests. Two help-center editing tools were paused over security and write-access questions, and an application-screening agent stayed local during security review. These are event outputs, not a measure of lasting adoption or lower maintenance cost. (Source)

Agent visibility was part of the build. The recap also describes Roo HQ, exposing agent history, reasoning, tool use, costs, and traces across work tools; Coach-a-Roo, for rating replayed support conversations; and frontend context and lint rules checked with two models. A proposal to organize scattered Claude skills was deliberately left for alignment. These were event artifacts and proposals, not verified long-term governance outcomes. (Source)

What They Do Well

• The honest documentation of where they diverged from prevailing norms. Most companies publish the idealized version of how they work. Buffer published why they chose NOT to be async-first. That specificity — “here is where we deliberately disagreed with the trend, and why” — is more useful than another generic remote-work manifesto.
• Salary transparency as an operating system. Buffer documents how its compensation approach evolves, rather than treating a formula as permanently settled. (Source)
• Shared financial outcomes. Inference: the approximately $5,000 average 2025 profit share may reinforce a connection between company performance and individual reward. The size of its cultural effect has not been measured here.
• Small team, mature product, sustainable economics. 2025 was their strongest year in years — Buffer’s 7th profit share post reports $2.51M net profit, with 15% redistributed to teammates (Source); the December 2025 shareholder update reports ~$1.95M MRR (≈$23.4M annualized) across ~69,760 paying customers (Source). Running a profitable bootstrapped-style company with 75 remote people is a different playbook than hypergrowth SaaS — and it works.

Tradeoffs and Weaknesses

The balanced sync/async approach requires judgment calls. Inference: Buffer’s routing guide leaves room for individuals to assess communication needs. A growing team may need more explicit response and escalation norms as interpretations diverge.

Concentrated staffing can create risk. Inference: a small function may be sensitive to a few departures. Buffer’s financial history records loss-making years, but it does not isolate staffing fragility as their cause.

Flexibility needs review. Inference: examine who regularly accommodates off-hours meetings and whether the burden is reciprocal. Buffer’s published norm does not establish that junior staff or contractors actually bear a disproportionate burden.

Limited public documentation depth. Compared to GitLab’s multi-thousand-page handbook or 37signals’ public guides, Buffer’s operating documentation is lighter. The blog posts are high-quality; the systematic operational documentation is thinner.

What Founders Can Copy

Inference — transfer guidance from Buffer’s dated case, not validated scale thresholds.

1. Explain the benchmarks and adjustments behind compensation, adapting them to your market.
2. Document why your communication choices fit the work.
3. Use the routing guide as a starting point, then specify response and escalation expectations.
4. Explain profit-share eligibility, allocation and dependence on company results.
5. Review who accommodates off-hours meetings and whether flexibility is reciprocal.

Where This Model Breaks

• When people interpret routing differently. Inference: review response and escalation norms as coordination demands change. The case does not establish a 200-person breakdown threshold.
• When profitability disappears. Buffer’s published history explicitly records no profit share in 2022 or 2023, both loss-making years. (Source) Inference: teams using profit share should explain this dependency before distributions become an expectation.
• When salary benchmarks lag market options. Inference: a published salary system still needs maintenance as roles and labor markets change. A location factor may create retention pressure where employees have alternatives at higher compensation; this is a potential risk, not a measured Buffer outcome.
• When sync availability becomes expectation. The “give flexibility to get flexibility” norm requires trust. If a manager starts scheduling mandatory early calls regularly without reciprocating flexibility, the norm collapses — and people in incompatible time zones bear the cost.

Related Practices

• Async Communication — Buffer’s honest documentation of when async fails is essential reading
• Cross-Time-Zone Collaboration — the “give flexibility to get flexibility” norm is a specific solution to this problem
• Decision Ownership — accountability and initiative in a distributed team

Related Frameworks

• Sync vs Async Decision Matrix — Buffer’s three-way routing framework is a direct input to this

Sources

1. Working at Buffer: https://buffer.com/resources/working-at-buffer/
2. Buffer is Remote but not Async-First: https://buffer.com/resources/remote-not-async-first/
3. Buffer Compensation Philosophy: https://buffer.com/resources/compensation-philosophy/
4. Reflecting on a Decade of Transparent Salaries: https://buffer.com/resources/salary-system/
5. Our 7th Profit Share — $377,005 Distributed: https://buffer.com/resources/7th-profit-share/
6. Buffer Shareholder Updates (primary source for monthly revenue / active-user / profitability data): https://buffer.com/shareholders
7. About Buffer (primary source for team count and team distribution): https://buffer.com/about
8. Join the Buffer Journey (current workweek and benefits): https://buffer.com/journey
9. February 2025 shareholder update (Linear adoption): https://buffer.com/shareholders/february-2025
10. December 2025 shareholder update (dated operating results): https://buffer.com/shareholders/december-2025
11. The Buffer Team Shipped 57 New Internal Tools in 5 Days (August 28, 2026): https://buffer.com/resources/build-week-2026/
12. We Send A Q&A Doc To Candidates Before We Meet Them (July 28, 2026): https://buffer.com/resources/buffer-q-a-doc/
13. We Added a $250 AI Tools Stipend for the Buffer Team (February 28, 2025): https://buffer.com/resources/ai-tools-stipend/
14. What We Learned After Finding 7 Forgotten Jobs Running for 5 Years (March 13, 2026): https://buffer.com/resources/infrastructure-refactoring/
15. How Buffer’s Content Team Uses AI (March 14, 2024): https://buffer.com/resources/how-we-use-ai-in-content/

Inferences

• Buffer’s guide combines tool routing with individual judgment. A team transferring it should review inconsistent interpretations; the available sources do not establish breakdown at any particular headcount.
• Salary transparency may reduce uncertainty about how pay is calculated, but public data does not establish that pay anxiety or disagreements disappear.
• Buffer’s recovery to profitability is a useful case for examining sustainable growth. Its financial history alone does not isolate team size or a particular value as the cause of that recovery.
• The AI stipend and Build Week widen who can ship an internal tool. The same recap shows that wider building immediately creates review work: some tools were paused for security or access. Event output does not establish that those tools stayed in use.

Work with Alex

If you are trying to design a communication system that is honest about what async can and cannot do for your team, Alex helps leadership teams build operating systems that match their actual context — not the idealized version.