Claude

Know the Business

Atlassian is a high-velocity software platform company that sells collaboration tools (Jira, Confluence, Jira Service Management) through a product-led flywheel rather than a traditional enterprise sales army. The business generates $5.2 billion in revenue at 83% gross margins and converts it to $1.4 billion in free cash flow, but the GAAP income statement is permanently distorted by $1.4 billion in annual stock-based compensation – making FCF the only honest measure of earnings power. The market is most likely underestimating the structural cost of the cloud migration wind-down and overestimating the durability of 20%+ revenue growth as the Data Center tailwind fades.

How This Business Actually Works

Atlassian's economic engine runs on three interlocking mechanics: land cheaply, embed deeply, expand relentlessly.

Landing costs almost nothing. Free tiers, transparent pricing, and self-service signup mean a 10-person startup can adopt Jira in minutes with a credit card. Over 90% of revenue each year comes from customers who existed the prior year, proving the flywheel works: existing accounts expand faster than new accounts land.

Embedding happens through workflow dependency. Jira becomes the system of record for engineering tasks, Confluence for documentation, JSM for IT tickets. Switching costs compound with every integration, automation rule, and institutional habit built on the platform.

Expanding is the profit lever. Atlassian grows within accounts by adding seats, selling premium/enterprise editions, cross-selling additional apps, and now bundling AI (Rovo) into Collections. Customers above $10K in Cloud ARR grew from 32,355 in FY2022 to 51,978 in FY2025, and these customers generate the majority of cloud revenue.

Cost structure is unusual for enterprise software: R&D is the largest expense at 51% of revenue (FY2025), while sales and marketing is only 22%. This is the product-led model in action. The downside: $1.36 billion of that R&D spend is stock-based compensation, not cash. The real cash cost of running the business is significantly lower than GAAP suggests, but the dilution is real.

The Server end-of-life (February 2024) forced customers to migrate to Cloud or Data Center, creating a one-time revenue tailwind. That tailwind is now largely exhausted. Cloud grew 28% in FY2025 but Data Center grew 21% – a material portion of "growth" is just migration, not organic demand.

The Playing Field

Three things jump out from this peer set.

First, Atlassian's GAAP operating margin is negative (-2%) despite $5.2 billion in revenue, while ServiceNow at $11.5B and Salesforce at $38B are solidly profitable on a GAAP basis. The gap is entirely SBC: strip it out and Atlassian runs at 25% non-GAAP operating margins. This is respectable but not exceptional – ServiceNow and Salesforce both achieve 30%+ on the same basis. Atlassian has more room to expand margins but is choosing to reinvest aggressively.

Second, Atlassian's FCF margin (27%) is in the same range as the best-in-class peers despite the GAAP losses. The cash conversion machine works. But FCF has plateaued: $1.416B in FY2024 vs. $1.416B in FY2025, essentially flat. That is a yellow flag for a company valued on growth.

Third, monday.com and Asana compete in the same work management space but are much smaller. Monday.com is growing faster (30%) with higher gross margins (89%) and is executing a credible enterprise land-and-expand of its own. This is the competitive threat Atlassian should be most worried about in the SMB/mid-market segment.

Microsoft is the existential risk. Azure DevOps, GitHub, Teams, and Planner collectively cover most of what Atlassian does. Microsoft bundles these into existing enterprise agreements at near-zero marginal cost. Atlassian survives because its tools are more specialized and beloved by developers, but the bundling threat is permanent.

Is This Business Cyclical?

Atlassian is low-cyclicality but not immune. Revenue growth decelerated from 26% (FY2023) to 20% (FY2025), partly from macro headwinds – management explicitly cited "expansion from existing customers moderate" among SMB/mid-market customers during the FY2024 slowdown.

The cycle hits Atlassian through three channels. Seat expansion freezes: when companies stop hiring or do layoffs, they stop adding Jira seats. This is the primary demand lever and it softens in downturns. Trade-down pressure: customers can move from Enterprise to Premium or Premium to Standard editions, compressing ARPU. New customer acquisition slows: fewer startups forming and fewer teams evaluating new tools.

However, churn is structurally low. Atlassian's tools become embedded in daily workflows, and the annual contract values are modest enough that they rarely get scrutinized in cost-cutting exercises. Revenue has never declined year-over-year in the company's history. In the 2022-2023 macro softening, revenue growth slowed from 34% to 26%, but absolute revenue and FCF both continued to rise.

The March 2026 layoff of 10% of the workforce (approximately 1,600 positions) was not driven by revenue decline but by a strategic reallocation toward AI and enterprise sales. The stock price decline from $242 (May 2025) to ~$65 (March 2026) – a 73% drawdown – was driven by multiple compression, not fundamental deterioration.

The Metrics That Actually Matter

Cloud ARR Customer Growth is the single best leading indicator. The 13% growth in >$10K Cloud ARR customers (45,842 to 51,978) in FY2025 signals that enterprise expansion is healthy but decelerating from the 18% pace in FY2024. Watch this number above all others.

SBC as a percentage of revenue is the metric most investors underweight. At 26% of revenue in FY2025, Atlassian's SBC is among the highest in enterprise software. The $1.36 billion in SBC exceeds the $1.42 billion in free cash flow – meaning on a fully-diluted, SBC-adjusted basis, the company is barely generating economic profit. FCF yield on the ~$17B market cap is roughly 8%, but SBC-adjusted FCF yield is near zero.

SBC has been growing faster than FCF. In FY2023, SBC exceeded FCF. By FY2025, they are roughly equal. If this trend continues, the "cash generation machine" narrative weakens significantly.

What I'd Tell a Young Analyst

Watch three things and ignore the noise.

First, the migration tailwind is ending. Server revenue hit zero in FY2025. Data Center grew 21%, but much of that was Server customers who landed in DC rather than Cloud. As this migration completes, the contribution to overall growth declines. The real question is whether organic Cloud growth (ex-migration) can sustain 20%+ on its own. Atlassian does not disclose this cleanly, and that opacity should concern you.

Second, SBC is the thesis killer hiding in plain sight. Atlassian trades at roughly 3x revenue and 12x FCF – seemingly cheap for a 20%-growth SaaS company. But if you subtract SBC from FCF, the "earnings" almost vanish. The share count has grown from ~253 million diluted shares (FY2022) to ~265 million (FY2025). Every year, the company gives away 26% of its revenue in stock to employees. The buyback program ($780M in FY2025) is mostly offsetting dilution, not returning capital. Ask yourself: is this a business that generates cash for shareholders, or one that converts customer payments into employee equity?

Third, Rovo AI is the optionality play, not the base case. Management claims 5 million monthly Rovo users and calls AI "the best thing to happen to Atlassian." If AI agents meaningfully improve the value proposition and justify higher pricing, Atlassian's installed base of 300,000+ customers becomes an enormous monetization opportunity. But AI features are table stakes across the industry – Microsoft Copilot, ServiceNow's Now Assist, and Salesforce Einstein are all pursuing the same thesis. Rovo needs to prove it can drive measurable revenue uplift, not just user engagement.