Atlassian

Their 'flexible project management' and 'knowledge, all in one place' positioning held beautifully when organizing work required dedicated software — but Claude can now sprint-plan, write tickets, and answer 'where did we document X?' without the subscription fee, the admin overhead, or the onboarding deck that nobody reads.

Project coordinators, confluence administrators, and service desk agents are discovering that ChatGPT can triage, prioritize, and route better than the workflow they spent six months configuring — and it doesn't need three approval layers to update a status.

The structural switching costs are real and formidable — years of ticket history, workflow integrations that touch every system, and teams who've muscle-memorized keyboard shortcuts — but that data lock-in only matters if the data itself remains valuable. When AI can generate project insights from a Slack export and three bullet points, the database becomes a very expensive CSV file.

The 'Rovo' rebrand and 'AI-powered apps driven by your team's knowledge' messaging suggests they understand the assignment, but shipping an AI feature that searches your existing Confluence pages faster isn't the same as reimagining why Confluence pages need to exist.

Teams still need to coordinate and track work, but the question 'should I create a ticket for this?' increasingly gets answered 'no' — when AI can manage context, dependencies, and handoffs in natural language, the elaborate project management ritual starts feeling like productivity cosplay.