Palantir's business model is built on the convergence of three distinct but interconnected revenue streams: government software contracts, commercial enterprise licensing, and — increasingly — AI platform subscriptions. Understanding how these streams interact, and why the company has deliberately kept them separate in its financial reporting, is essential to understanding Palantir's long-term strategic logic. The government segment, anchored by the Gotham platform, operates on multi-year contracts with defense and intelligence agencies. These contracts are often structured as Time-and-Materials or Cost-Plus arrangements, which provide revenue visibility but traditionally compressed margins. Palantir has worked to shift these toward fixed-price software licenses, which carry dramatically higher gross margins. Government contracts are not merely revenue — they are reference deployments that validate Palantir's capabilities in the most demanding analytical environments on earth. The U.S. Army, the Department of Defense, the NHS, and NATO allied forces represent anchor clients that create a halo of credibility impossible to manufacture through conventional enterprise sales. The commercial segment, powered by Foundry and increasingly by AIP, operates on a subscription-and-expansion model. The unit economics are structured around a land-and-expand strategy: initial deployments are scoped conservatively to ensure rapid time-to-value, and expansion is driven by demonstrated ROI. Palantir tracks a metric called "Top 20 customer revenue" to illustrate how its best commercial relationships compound over time. When a customer deploys Foundry across one business unit and achieves measurable efficiency gains, the natural motion is to expand to adjacent units — each expansion increasing average contract value without proportional increases in cost-to-serve. The AIP bootcamp model has introduced a third dynamic: velocity. Traditional enterprise software sales are slow, relationship-driven processes. AIP bootcamps bypass the lengthy RFP-and-evaluation phase by demonstrating value in real time. Customers arrive with their own data, work with Palantir engineers for three to five days, and leave with a functional prototype. The psychological and commercial effect is powerful: customers who have already built something with AIP are far more likely to sign a contract than those who have only seen a demo. This model is also highly scalable — Palantir has conducted hundreds of bootcamps globally, and each one generates both pipeline and product feedback. Palantir's pricing model reflects its positioning as a premium, strategic vendor rather than a commodity data tool. Annual contract values for mid-market commercial customers typically range from $1 million to $5 million, while strategic government contracts can exceed $100 million per year. The company does not compete on price — it competes on the depth of integration, the sophistication of the ontological data model, and the operational outcomes delivered. This pricing posture naturally limits the addressable market to large organizations with complex data problems and the organizational maturity to deploy the platforms effectively. The company's cost structure is dominated by research and development and sales and marketing, both of which are investments in future revenue rather than current-period costs. Palantir historically ran at a loss because it chose to invest aggressively in platform development and forward-deployed engineers — a model it borrowed from management consulting, where senior talent is embedded with clients to drive adoption. This forward-deployment model creates stickiness that pure SaaS vendors cannot replicate: when a Palantir engineer has spent six months inside a client's operations, the relationship and institutional knowledge embedded in the deployment make switching practically and organizationally costly. The transition to GAAP profitability in 2023 and sustained profitability through 2024 signals a maturation of the business model. Palantir has not abandoned its investment posture, but it has reached a scale at which revenue growth is outpacing operating expense growth — the classic inflection point for high-margin software businesses. With gross margins consistently above 80%, the operating leverage embedded in Palantir's model suggests that continued revenue growth will translate disproportionately into operating income.