(Reuters) — Collaboration software provider Slack Technologies Inc is building a safety engineering team that will develop methods to help reduce the disruptions that have been more frequent on Slack’s service than rival systems, the company told Reuters.
Recurring outages have prevented Slack users from connecting with its service as the fast-growing startup competes against the likes of Microsoft, Alphabet’s Google, Cisco Systems, and Facebook in the crowded workplace collaboration software market.
Since April 2017, Slack has suffered 40 days with outages as well as 51 days where some users suffer incidents that reduce functionality, according to the company’s status website.
In the same time frame, rival Atlassian’s services Hipchat and Stride have incurred a combined four days with outages and 25 days with incidents, according to Atlassian’s status websites. Competing services Microsoft Teams, Google Hangouts Chat, Workplace by Facebook and Cisco Spark do not provide comparable service status statistics.
Privately held Slack launched its workplace collaboration software in 2014 and now has a base of more than 6 million daily active users.
Slack will task the safety engineering team with discovering “failure modes before they impact service to customers or cascade into full outages,” according to a job listing posted for the lead manager.
Slack communicates its service issues “in great detail” to customers, and “maintains a consistently high uptime, the industry standard for reliability, routinely hitting 99.99 percent or better,” the company said in a statement to Reuters.
“As our customer base grows, we are dedicated to doing two things very well: continuing to provide a service that people love to use; and ensuring the uptime and reliability of our product consistently improves,” Slack’s statement said.
Continued outages carry risks for the company, said David VanAmburg, managing director at the American Customer Satisfaction Index, which analyzes customer satisfaction for clients.
“Customers expect reliability,” VanAmburg said. “The company needs to determine how to update its infrastructure or risk losing the good will of its users.”
The job listing said Slack’s team will use a new technique known as chaos engineering, which involves controlled stress testing of systems to identify issues, said Kolton Andrus, chief executive of Gremlin, a startup that builds software for such testing.
“Slack is going through those growing pains,” Andrus said. “The software that got them through the early days of the company needs to either be rewritten or grow.”