Claude

The Full Story

Atlassian spent three years executing the most consequential platform transition in its history: killing its on-premise Server business, migrating customers to Cloud and Data Center, and then pivoting hard into AI as the next growth engine. Revenue nearly tripled from $2.1 billion in FY2021 to $5.2 billion in FY2025, yet the stock lost roughly 65% of its value between mid-2025 and early 2026 as investors repriced every software company against AI disruption risk. Management credibility on operations has been strong – they consistently beat revenue guidance – but the narrative has shifted from "inevitable cloud migration compounder" to "prove AI is additive, not destructive."

The Narrative Arc

The narrative arc has three distinct chapters. Chapter 1 (FY2021-FY2023) was the Server-to-Cloud migration story – management bet the company on forcing 260,000+ customers off legacy on-premise deployments, enduring short-term revenue headwinds from the maintenance decline for the long-term prize of recurring cloud subscriptions. Chapter 2 (FY2024) was the harvest year – Server support ended, migration tailwinds boosted Data Center revenue 48%, and FCF nearly doubled. Chapter 3 (FY2025-present) marks the AI pivot, where management reframed Atlassian from a collaboration tools company into a "system of work" platform with AI agents at the center.

Each chapter required investors to accept a new thesis. The current chapter – AI as value creator rather than value destroyer – remains unproven.

The stacked chart tells the migration story visually. Server revenue collapsed from $608M to zero in four years. Cloud revenue filled the gap and then some, growing from $968M to $3.4B. Data Center – originally positioned as a transitional product – became a surprisingly durable $1.5B business for customers who need on-premise.

What Management Emphasized – and Then Stopped Emphasizing

Several shifts stand out:

Server migration: from centerpiece to footnote. In FY2023, nearly every investor communication centered on migration tooling, pricing incentives, and the February 2024 deadline. By FY2025, "Server" barely appears in the 10-K except as a line item going to zero.

AI/Rovo/Agents: from zero to everything. AI was mentioned only in the context of "artificial intelligence capabilities" in FY2023's platform description. By FY2025, Rovo became a named platform app with three sub-products (Search, Chat, Studio), AI agents were framed as "virtual teammates," and the entire product portfolio was reorganized from "products" to "apps, agents, and Collections." This is the fastest narrative pivot in Atlassian's history.

Customer count: quietly deprioritized. In FY2023, management highlighted "more than 260,000 customers" prominently and tracked total customer count as a key business metric. By FY2024, the key metric shifted to "customers with greater than $10,000 in Cloud ARR" – a higher-quality metric, but also one that narrows the discussion away from total customer growth. By FY2025, the total customer count was mentioned once as "more than 300,000" without the precision of earlier years.

Enterprise sales: growing in prominence. The FY2023 10-K described sales as primarily automated and low-touch. By FY2025, management introduced language about "field sales, solution sales" and explicitly acknowledged the need for a "dedicated sales team" to expand enterprise relationships. Atlassian is quietly becoming more like the traditional enterprise software companies it once defined itself against.

Risk Evolution

The risk landscape has fundamentally shifted. Three years ago, the primary risk was whether customers would successfully migrate off Server before the deadline. That risk resolved – it worked. Today, the dominant risk is existential: will AI-powered coding agents and automation tools reduce demand for the project management and collaboration software that is Atlassian's core business?

The February 2026 Bloomberg headline – "Atlassian Founders Lose $7.2 Billion as Software Stocks Slump on AI Fears" – crystallized what the market had been pricing in through late 2025. The stock fell from approximately $240 in mid-2025 to below $65 by April 2026, a decline driven not by any operational miss but by a sector-wide reassessment of software value in an AI world.

Stock-based compensation has become increasingly material. SBC grew from $341M in FY2021 to $1.36B in FY2025, representing 26% of revenue. This is the reason GAAP operating losses persist despite strong non-GAAP margins – and it creates genuine dilution risk even as the company executes buybacks.

The March 2026 restructuring (approximately 1,600 employees, or 10% of workforce) is the second major workforce reduction in three years. CEO Mike Cannon-Brookes framed it as reshaping the workforce for AI. The departure of the CTO alongside this restructuring raised questions about whether this was strategic repositioning or reactive cost-cutting.

How They Handled Bad News

Atlassian's management communication pattern during setbacks follows a consistent formula: acknowledge the headline, reframe the narrative toward long-term opportunity, and highlight metrics that remain strong.

The FY2023 layoff (500 employees, ~5%): Management described this as "rebalancing resources" to "accelerate progress against our largest growth opportunities." The restructuring charges totaled $97M. In the same period, they authorized a $1B share buyback – a signal meant to say "we believe in the stock even as we cut costs."

The FY2025 guidance disappointment (Q3 FY2025, November 2025): Revenue came in at $1.43B, up 20.6% YoY, beating estimates. But full-year guidance disappointed investors. The stock traded flat despite the beat – a tell that the market was no longer rewarding operational execution and was instead fixated on the AI disruption thesis.

The March 2026 restructuring (1,600 employees, ~10%): This was a larger restructuring than the first. Management took $225-236M in charges and simultaneously announced the CTO's departure. CEO Cannon-Brookes told employees AI "changes the mix of skills" and "the number of roles required in certain areas." By coupling a restructuring with an AI narrative, management tried to position the cuts as forward-looking rather than defensive. The market was not persuaded – the stock fell further.

Guidance Track Record

Score: 8 out of 10. Atlassian's operational guidance track record is excellent. They have consistently met or exceeded revenue, margin, and cash flow targets across three fiscal years. The deductions come from two areas: (1) the shift in key metric from total customers to Cloud ARR >$10K customers was never explicitly framed as a metric change, which obscures deceleration in the broader customer base; and (2) the Q3 FY2025 full-year guidance disappointed the market, suggesting management's confidence signals were not as strong as investors needed during a period of sector-wide doubt. Still, when management makes a specific numerical promise, they deliver.

What the Story Is Now

The current story rests on three pillars, each with a different level of de-risking:

Pillar 1: The cloud migration is complete and successful. This is fully de-risked. Server revenue is zero. Cloud revenue is $3.4B and growing 28%. Data Center provides a $1.5B bridge for customers who need on-premise. The migration forced customers to make a decision, and most stayed.

Pillar 2: The enterprise expansion is working. Partially de-risked. Cloud ARR customers above $10K grew from 32,355 to 51,978 over three years. RPO grew 44% to $3.8B in Q2 FY2026, indicating larger, longer-term contracts. The hiring of a Chief Revenue Officer in late 2024 and the new "Collections" packaging (Teamwork Collection, Strategy Collection) signal a real push upmarket. But marketing and sales expense grew 29% in FY2025, faster than revenue, which suggests this motion is not yet as efficient as the flywheel that built the company.

Pillar 3: AI will be additive, not destructive. This is the core unresolved question. Atlassian is betting that AI agents will work inside Jira and Confluence, making those platforms more valuable rather than making them obsolete. Rovo reached 5 million monthly active users. But the market is pricing in the possibility that AI coding agents could reduce the number of software developers – which would directly shrink Atlassian's seat-based revenue model.

The margin chart captures the central tension. Non-GAAP operating margin has expanded steadily from 20% to 25%, demonstrating genuine operating leverage. But GAAP operating margin remains negative, stuck around -3%, because SBC consumes the entire improvement. This divergence is not sustainable forever – eventually, the market will demand that the GAAP picture converges toward the non-GAAP one.

What to believe: Revenue execution, margin expansion, and cash generation are real and well-documented. FCF of $1.4B on $5.2B in revenue is a 27% margin. The cloud migration was successful, and the enterprise motion is gaining traction.

What to discount: The AI narrative is still aspirational. Rovo adoption metrics (5M MAUs) are early and lack revenue attribution. The SBC problem is significant – $1.36B in FY2025 means FCF minus SBC is effectively zero. The "system of work" positioning is marketing language that has yet to translate into a defensible moat against AI-native competitors.

What has changed: Atlassian is no longer the scrappy, low-touch, anti-enterprise company it was founded as. It now has a dedicated sales force, enterprise Collections, a CRO, and marketing spend growing faster than revenue. This is not necessarily bad – it may be necessary – but it represents a fundamental shift in the company's identity that long-term investors should acknowledge.