Slack is building a safety engineering team that will develop methods to help reduce outages; since April 2017, Slack has had 40 days with outages
Salvador Rodriguez / Reuters : Tweets: @patrickmoorhead . Thanks: @sal19 Tweets: Patrick Moorhead / @patrickmoorhead : Like the Twitter “fail-whale”, this must get fixed soon. I can't imagine any large or medium sized enterprise accepting this at all. #FutureofWork http://twitter.com/... Thanks: @sal19
Context & Ripple Effects
The reliability push lands mid-pivot: months earlier Slack had launched Enterprise Grid specifically to win large firms with unlimited workspaces and admin controls, but Moorhead's comparison to Twitter's fail-whale era captures the risk — enterprises won't standardize on a collaboration layer that goes dark. The Reuters report puts a number on the problem: 40 days with outages since April 2017.
The coverage that follows suggests the team was necessary but not sufficient — Slack went on to suffer a six-hour outage in October 2020, repeated degraded performance in 2022, and another full-app outage in 2023, while analysts tied its standalone struggles to distribution rather than product.
First-order effects
- Slack's enterprise sales teams now have a concrete answer to the reliability objection blocking Enterprise Grid deals, but also a public admission that outages were frequent enough to warrant a dedicated engineering group.
Second-order effects
- Vendors selling failure-simulation tooling gain a marquee validation point — Gremlin's $18M Series B later that year rode the same premise that services should rehearse outages before customers live them.
Third-order effects
- Reliability shifts from an ops afterthought to a procurement gate for workplace software: if a company's messaging layer fails, work stops, so buyers increasingly treat uptime commitments as contract terms — though Slack's recurring 2020–2023 outages show a dedicated team alone doesn't close the gap.
The trend: Collaboration platforms are being forced to treat reliability as core infrastructure investment as they move upmarket into enterprises, with chaos-engineering tooling emerging as the supporting industry.