The Whole Product Framework

A model alone only wins innovators; each step from generic to expected, augmented and differentiated product wins the next customer segment, and costs more.

Contents
    Four stages of a product drawn as flowers, from a lone Models/Intelligence core to a core ringed by petals (context, gateway, ops, UX, legal, agents) with integrations at the edges, mapped to innovators, early adopters, early majority and late majority, with the chasm between the second and third. A scale at the bottom weighs simple, cheap and fast against guarded, compliant and complex.

    Use it when Deciding what has to surround your model before the next customer segment will adopt it.

    A model on its own is a generic product. Each of four stages wins one customer segment from Moore’s Crossing the Chasm.

    • Generic product, for innovators (techies). The enabling technology: “a barebones model appealing to innovators/techies who prioritize core functionality.”
    • Expected product, for early adopters (visionaries). “The minimum that needs to be there to do the job”: context and knowledge, a gateway and ops around the model, enough for an MVP.
    • Augmented product, for the early majority (pragmatists). Multiple subsystems (UX, legal, agents, ops, gateway) form a system that solves the problem, shaped by constraints such as latency, reliability, cost, safety and security. Reaching them means crossing the chasm.
    • Differentiated product, for the late majority (conservatives). Integrations and niche customizations at the edges are “what makes the product differentiated, what creates a brand, and what makes customers stick.”

    Every step is a tradeoff. The early end is simple, low cost and low latency, but has no guardrails, observability or compliance. The late end is guarded, targeted, compliant, safe and sustainable, but slower, more expensive and more complex. The question underneath never changes: what problem are you solving?

    Topics

    Agents in Production 10Building with AI 9AI Product Strategy 24

    Start here

    How Agents Run in ProductionBeyond LLMs: The Whole Product Framework for AIThe Verification Bottleneck: Why AI's Real Cost Is Human Attention

    ↑ ↓ to move, Enter for all results