2 ms·
Six months ago they announced plans to grow their headcount in India by 1500 before end-2024. Their worldwide headcount growth since 2019 [0] does feel, in ret
by Jedd 4y ago
Six months ago they announced plans to grow their headcount in India by 1500 before end-2024.
Their worldwide headcount growth since 2019 [0] does feel, in retrospect at least, untenable - year on year - 3600, 4900, 6400, 8800. Arguably this fits into the same 'correction' category as other big players in recent months.
Reading the potted history [1] of acquisitions / sell-offs is a trip down memory lane. It feels like even their big-name acquisitions are still not terribly well integrated -- and combined with their famous lack of a 'traditional sales team', it does invite some obvious, if churlish questions.
[0] https://www.statista.com/statistics/1276817/atlassian-number-employees/ https://www.statista.com/statistics/1276817/atlassian-number...
[1] https://en.wikipedia.org/wiki/Atlassian#Acquisitions_and_product_announcements https://en.wikipedia.org/wiki/Atlassian#Acquisitions_and_pro...
- 908B64B197 4y ago> Six months ago they announced plans to grow their headcount in India by 1500 before end-2024. I know a lot of people moving away from Atlassian because of performance issue. Not sure how this will reverse the trend.
- arlcode 4y agoFor us at least their "forced" switch to Jira cloud was a massive increase in cost (which was the intention I guess). Their developer offering at bitbucket is also more expensive and lacks critical features which are supported by the competition. The market is bound to react somehow. On the other hand vendor lock in is a powerful tool for atlassian.