The Remote-First Operating Stack
Five mechanisms that make an all-remote company across 60+ countries actually work
- Difficulty
- Advanced
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 93%
GitLab is the world's largest all-remote company (2,000+ people, 60+ countries) and has been remote-first since long before the pandemic. DeSanto compresses what makes it work into a small, concrete stack: transparency, results-not-hours, deliberate over-communication at 150%, asynchronous-first decision-making with an empowered DRI, and periodic in-person gatherings — plus one behavioural rule ('don't wait') that prevents multi-day blocking cycles.
Origin
David DeSanto's synthesis of GitLab's remote practice, delivered as the advice he gives leaders who reach out to him on LinkedIn and X. The 150% communication heuristic comes from his executive coach.
Core principles
- 01Be transparent — remote work forces specificity in communication, and transparency is what makes that specificity available to everyone.
- 02Focus on results, not hours — the 40-hour question is the wrong question in every direction.
- 03If you think you're communicating 100% accurately, the other person received 60-70%; aim for 150%.
- 04Async-first: key decisions that involve someone in a non-overlapping time zone wait for that DRI to come online.
- 05Meetings are optional, recorded, and readable so a different time zone can consume them.
- 06Empowerment is the time-zone strategy — you cannot cover every call, so give people the authority to decide without you.
- 07Human connection is a real input — periodic in-person events make Zoom feel less scary.
How to run it
- 1
Set async as the default decision mode
Key decisions involving the directly responsible individual (DRI) wait for that person to come online rather than being made by whoever happens to be awake. Meetings are optional, recorded, and have written readouts good enough for a different time zone to consume.
Pro tip Name a DRI for every decision. Async only works when it is unambiguous whose call it is.
Watch out Deciding without the absent DRI once teaches everyone in that time zone that they must attend meetings at 3am to stay relevant.
- 2
Overcommunicate to 150%
Assume your message landed at 60-70% of what you intended, and deliberately over-explain to compensate. In writing, this means the requirement or decision is stated more completely than feels necessary.
Pro tip Applies hardest to written requirements — see the 'don't wait' rule for the correction mechanism when 150% still isn't enough.
- 3
Install the 'don't wait' rule
If you suspect something is going off the rails, or you just want to know it's on track, do not wait for the weekly standup. Ask on the issue, comment on the merge request, or send a Slack message immediately. Letting a 24/48/72-hour cycle pass with an engineer blocked or a question unanswered is the failure.
Pro tip Pair this with the short-toes value so the check-in reads as unblocking, not as distrust.
Watch out Without the kindness values in place, high-frequency check-ins read as micromanagement and get resented.
- 4
Empower across time zones instead of covering them
Accept that leaders will have minimal overlap with parts of their team and solve it with delegation, not attendance. DeSanto had a leader in Tel Aviv with a half-hour daily overlap who ran a globally dispersed and highly effective team; he has West Coast leaders empowered to take APAC-friendly calls in his place.
Pro tip If a leader gives clear outcomes and communicates asynchronously, zero overlap is genuinely survivable.
- 5
Buy back human connection deliberately
Schedule in-person gatherings at whatever cadence you can afford: leadership team every quarter, directors-plus every other quarter, division-level meetups. Don't wait for the company-wide offsite.
Pro tip DeSanto got all 140 people in his product division together across two cities over the course of a month rather than waiting years for a single all-company event.
Watch out Remote genuinely is not for everyone — DeSanto says the most common reason a strong hire doesn't work out at GitLab is missing the in-office feeling, not underperformance.
In the wild
DeSanto had a direct report based in Tel Aviv while he lived in the middle of the United States — a thirty-minute daily overlap. Her own team was also geographically dispersed.
→ She was consistently one of his most efficient leaders and her team was highly effective, because she gave clear outcomes and empowered her people to decide without her.
Rather than waiting for GitLab's first company-wide gathering since the pandemic, DeSanto assembled all 140 people in the product division across two different cities over the course of a month.
→ People finally met each other in person, which he says makes subsequent Zoom interaction 'a little less scary' because the human connection already exists.
Common mistakes
Policing hours in either direction
DeSanto notes people fixate on both 'did you work the full 40 hours?' and 'wow, you worked more than 40 hours, what's going on?'. Both questions substitute a proxy for the agreed outcome and both corrode trust in a remote setting.
Waiting for the weekly standup to raise a concern
In a co-located office you tapped someone on the shoulder. Remote, that instinct must be consciously converted into an immediate issue comment or Slack message — otherwise a blocked engineer stays blocked for days.
Treating in-person events as optional nice-to-haves
The connection deficit is the single most common reason good people leave an all-remote company. Budgeting for periodic gatherings is retention spend, not perks spend.
Is it for you?
Best for
Leaders scaling a distributed or all-remote org past the point where informal coordination works — especially across many time zones.
Not ideal for
Teams whose work genuinely requires physical co-presence, and individuals who need in-office human contact to be happy (a fit problem, not a process problem).
From the transcript
“things like be transparent we've talked about that a bunch already uh focus on results not hours”
“if you think you're communicating 100% accurately that's probably 60 or 70% for the other person and shoot for like 150%”
“we focus on asynchronous communication first”
“if you're concerned or wondering something is going off the rails or you just want to know something's a on track like don't wait until…”
From the episode
The GitLab way: Kindness, transparency, and short toes
David DeSanto (CPO)