Salesforce Ecosystem Decomposition

In the AI Story

The decomposition is granular because the moats are heterogeneous. The data layer includes fifteen-plus years of customer interactions, deal histories, and pipeline records specific to each enterprise customer — data that cannot be replicated by writing new code because it was generated by the actual operation of the business over time. The integration layer includes hundreds of certified connections to enterprise systems (marketing automation, ERP, billing, business intelligence) maintained over years. The marketplace layer includes thousands of third-party applications on the AppExchange whose developers have aligned economic interests with Salesforce's continued existence. The trust layer includes SOC 2 Type II, FedRAMP, HIPAA, and GDPR certifications obtained through years of process maturity. The human capital layer includes millions of Salesforce-certified administrators, developers, consultants, and architects whose careers depend on the platform.

Each layer requires different valuation treatment. Data assets compound — each year of additional customer interaction data makes the layer more valuable, more unique, more irreplaceable. Integration density compounds — each new customer deployment deepens the institutional knowledge of how to integrate Salesforce with adjacent systems. Marketplace network effects produce increasing returns. Trust accumulates only through demonstrated performance and cannot be accelerated. Talent ecosystems self-reinforce. The aggregate effect is that Salesforce's ecosystem is not merely large; it is structurally durable in ways that the code-devaluation narrative does not threaten and may strengthen.

The valuation discipline mirrors the broader Damodaran method. The code component is valued with: revenue growth at 3-5%, operating margins compressing from 22% to 15% over a decade, a 12% discount rate (CAPM plus a 2-3 point disruption premium), and a finite-life terminal value reflecting twenty-year horizon. The ecosystem component is valued with: revenue growth at 7-10%, margins stable at 22-25%, a 9% discount rate (CAPM minus a 1-2 point durability discount), and perpetuity terminal value with a 2.5% terminal growth rate. The arithmetic is mechanical; the judgment is in the assumptions.

Origin

The decomposition emerged in Damodaran's commentary on the SaaSpocalypse in February-March 2026, building on his decades of work on enterprise software valuation and his observation that the market's uniform repricing failed to distinguish between Salesforce's code value and its ecosystem value.

Key Ideas

Sum-of-parts reveals what the multiple conceals. The aggregate revenue multiple (six to seven times revenue post-correction) hides the fact that the market is implicitly assigning zero value to the code-dependent business.

Five layers of ecosystem. Data, integrations, marketplace, trust, and human capital each contribute to the moat and each require their own valuation treatment.

Sensitivity to ecosystem share. If ecosystem revenue is 75% rather than 65%, intrinsic value rises by 15-20%; the assumption is the most consequential single judgment.

Moderate undervaluation, not screaming buy. The 20-25% gap is meaningful but requires patience and willingness to revise as evidence accumulates.