---
id: ins_software-is-not-a-moat
operator: Evan Spiegel
operator_role: CEO and co-founder, Snap
source_url: https://www.lennysnewsletter.com/p/snapchat-ceo-why-distribution-is
source_type: podcast
source_title: Evan Spiegel — Distribution, design as bottleneck — Lenny's Podcast
source_date: 2026-04-28
captured_date: 2026-05-01
domain: [founder-craft, ai-native, gtm]
lifecycle: [strategy-bets]
maturity: foundational
artifact_class: framework
score: { originality: 4, specificity: 4, evidence: 5, transferability: 5, source: 5 }
tier: A
related: [ins_jtbd-as-agent-wiring-diagram, ins_design-as-bottleneck]
raw_ref: raw/podcasts/evan-spiegel--snap-distribution-design-bottleneck--2026-04-28.md
---

# Software is not a moat, ecosystems, hardware, and distribution are

## Claim
Snap learned 15 years ago that software features are copy-paste. The durable moats are ecosystems (creators, AR developers, lens platform), hardware (Spectacles), and distribution. The same lesson is now hitting AI: prompts and models commoditize. The defensible asset is the team-shaped infrastructure, the relationships, and the data flywheel underneath.

## Mechanism
Software is reproducible at near-zero cost; the marginal cost of copying a feature is engineering time. Once a feature works, multiple competitors ship it within weeks. Ecosystems and hardware have switching costs and network effects that engineering time cannot replicate. Distribution requires channel relationships, brand trust, and platform leverage that compound over years. The moat is whatever the next competitor can't copy in a quarter.

## Conditions
Holds when:
- The category has clear copy-paste mechanics (mobile apps, SaaS UI, prompts).
- The team has access to ecosystem-builder, hardware, or distribution leverage.

Fails when:
- The "ecosystem" is theatrical, a developer portal nobody uses doesn't moat anything.
- The product is genuinely novel and the moat *is* the technical implementation. Rare but real (early Stripe, early Cursor).

## Evidence
> "15 years ago, we essentially learned that software is not a moat, which is something that everyone is discovering today with AI."

Snap's moats: AR developer ecosystem, lens platform, creator relationships, Spectacles hardware, distribution through close-friend networks. Each compounds; each is hard to copy.

· Evan Spiegel on Lenny's Podcast, 2026-04-28

## Signals
- Investment is allocated toward ecosystem and platform work, not just feature parity.
- Competitors copy your features quickly without making a dent in your share.
- Internal language shifts from "what should we build" to "what loops should we deepen."

## Counter-evidence
Bret Taylor's outcomes-pricing thesis suggests AI agents will change moat economics again, outcome delivery may become a new defensible layer. Krithika Shankarraman flags taste and proprietary data as the AI-era moats, more granular than "ecosystem." All three are pointing at the same shape: durable assets that survive copy-paste.

## Cross-references
- `ins_design-as-bottleneck`, Snap's specific implementation of the cohesion mechanism
- `ins_jtbd-as-agent-wiring-diagram`, the AI-era equivalent moat-building work