Roughly seventy percent of what your people know is never documented.

Every industry with a trained, distributed workforce holds the same hidden asset. The workforce knows things the corporate center cannot see. None of the existing instruments capture it at scale (surveys are too shallow, focus groups are too small, human interviews are too expensive, knowledge management tools assume the knowledge has already been written down). The buyer for KAAS already spends in the multi-million-dollar range per market on research, R&D, and consulting that targets the layer above this one. KAAS delivers more useful intelligence at one to two percent of that cost, and it delivers it as a business unit rather than as a study.

The first vertical archetypes the same playbook runs against are large retail with a brand-partner ecosystem, labor-intensive multi-unit operations, and senior expertise services. Each one is described below in terms of the shape of the opportunity and the monetization pattern that fits it. The adjacent verticals behind them are described last.

A large beauty-retail network can hold tens of thousands of advisors at the counter across many markets. Each advisor sees dozens of customers a day. The aggregate is well over a million customer interactions a day. None of what those advisors observe ever reaches the brand partners on the shelf. The knowledge is in the advisor's head and dies at the end of every shift.

The adjacent AI-in-beauty market is multi-billion dollar and growing at high double-digit CAGR through the rest of the decade. The KAAS pricing works because the buyer is purchasing infrastructure with three endpoints. A research report prices in the low six figures. A SaaS seat prices per user per month. A layer that monetizes the brand partners on the shelf through a scoped API is priced against the downstream revenue it enables, with an enterprise floor. Capturing a small percentage of the adjacent AI spend lands the deal in a band that no tooling pricing model can reach.

Why beauty retail goes first. The category is built around a small number of operators who already monetize brand partners directly through shelf space, which means the second revenue endpoint is a contract change rather than a new sales motion. The first KAAS client in the vertical arrives with a captive distribution layer.

A multi-brand restaurant operator can run thousands of crew across its restaurant brands. The workforce knows exactly what the crews do per shift, task by task, hour by hour. The corporate center cannot see that level of detail from any current system. The wedge is a productized labor intelligence layer with AI-readiness assessment per role.

At three cents per enriched conversation, the capture economics scale cleanly. A thousand-conversation tier costs about thirty dollars in compute. A workforce-wide capture across tens of thousands of crew costs a four-figure compute bill against a business-unit-tier deal. The capture stage is essentially free at scale. The value compounds downstream.

The same playbook runs across the rest of the QSR category and across casual-dining chains of similar scale. The brand-partner endpoint at the restaurant tier is the food and beverage supplier ecosystem. The suppliers behind the menu have a similar appetite for floor-level intelligence on AI rollout and shift economics.

A major consultancy can hold thousands of senior consultants on a regional bench. Each one knows what actually works on a large enterprise deployment, as opposed to what is written in the playbook. The wedge is the productization of that expertise into a layer the firm licenses to its own enterprise clients. The deliverable shifts from billable hours to a continuously refreshed capability map per practice.

The vertical-specific ten-dimension schema replaces two or three dimensions relative to retail. For consulting, the brand-perception map is replaced by engagement-economics tells, and the customer-segment behavior dimension is replaced by client-archetype patterns. The other seven dimensions hold.

The expansion path runs up to the largest global integrators and down into regional advisory practices. Each one has the same structural blocker that keeps it from building KAAS internally: the billable-hours model is incompatible with productized infrastructure. Decoupling the methodology from the partners directly cannibalizes the consulting business.

Five tiers, each anchored on capture scale. The right comparable is the enterprise client's current research and R&D budget per market (typically in the multi-million-dollar range), and the cost of a survey is a category error. We deliver more useful intelligence at one to two percent of that comparable, and we deliver it as a business unit. The conversation with the buyer is about the business unit being built. The pricing artifact is structured around capture scale.

• PulseEntry tier

  A few hundred conversations

  Proof the workflow works. The first node.

• MacroProduction tier

  A full regional market mapped

  A regional market mapped end to end. The standard production deployment.

• RefreshCadence add-on

  Cadence add-on

  Keeps the layer fresh on a recurring schedule.

• PartielDeep-dive tier

  Long-form interviews for complex domains

  Forty-five-minute conversations for senior expertise and high-complexity verticals.

• Custom EnterpriseEnterprise

  Multi-market

  Brand-partner licensing infrastructure plus AI-agent API distribution.

We do not discount below the entry-tier floor. A token pilot signals tooling. An entry-tier engagement signals a category move. Holding the floor preserves the positioning. Crossing it destroys the next ten conversations at infrastructure pricing.

Once the first vertical archetypes have shipped, the same playbook runs against six adjacent verticals. Each one has the same fundamental gap: a trained, distributed workforce holding strategic understanding the corporate center never sees.

• Healthcare delivery

  Hundreds of thousands of pharmacists per chain network

  Clinical floor signal and drug-interaction observation. Patient concern recurrence patterns.

• Legal services

  Partner-level expertise at the largest law firms and advisory benches

  Matter-pattern intelligence. Engagement renewal predictors. Cross-jurisdiction practice patterns.

• Financial services

  Tens of thousands of private-wealth advisors per major brokerage

  Client-conversation knowledge the firm has never structured. Cross-segment behavioral patterns.

• Energy and industrial

  Oil and gas field operations: rigs, refineries, substations

  Operator-level intelligence that never reaches the C-suite. Safety pattern recognition at the asset level.

• QSR and restaurant networks

  National multi-brand quick-service and casual dining chains

  Task-level labor intelligence. AI-readiness assessment per role. Shift-level operational variance.

• Pharmacy and drugstore chains

  National pharmacy chains at the hundred-thousand-pharmacist scale

  Prescription friction signal. Patient concern aggregation. Cross-store practice variance.

Total addressable market across these verticals: tens of billions in current market research, R&D, and consulting spend, none of which captures the organic brain layer KAAS captures. The capacity constraint is not market demand. It is the vertical-by-vertical schema work, the methodology calibration, and the brand-partner economic model that each vertical requires before scale. Five verticals shipped poorly is worse than one vertical shipped well.

Every Fortune 500 is building AI agents for customers, employees, and operations. Every AI agent needs verified domain ground truth. The grounding data is currently coming from three places, each of which fails for a specific reason:

• Scraped from the web.

  Legally risky, factually dated, contains the same hallucinations the agent is trying to eliminate.

• Synthesized by other LLMs.

  Factually unreliable, recursive, training-data contaminated. The synthetic respondent cannot surface anything the model did not already know.

• Bought from data brokers.

  Sourced from a third party, available to every competitor on the same terms. Captures exhaust rather than intent.

KAAS is the verified source. Real workers, structured interviews, captured on cadence, enriched for AI consumption. When a brand partner builds its next AI customer agent, that agent needs floor truth for the vertical it serves. The brand can build it internally (expensive, slow, conflict-of-interest). Or it can license it from the proprietary layer the retailer already owns. That license is the second revenue line and the long-duration moat. Once the brand's AI agents are wired to consume the layer, switching cost compounds with every new agent task.

Most adjacent categories are blocked from KAAS by a structural reason that retrofitting cannot solve. Qualtrics cannot reach the brand-partner endpoint because its install base is HR and CX. Medallia cannot repurpose customer VoC data for brand-partner monetization because consent was captured for one purpose and cannot legally be repurposed in most jurisdictions. McKinsey cannot productize without cannibalizing the consulting wrap. D&B cannot build the AI-interview muscle fast enough to compete with KAAS-first players for senior qualitative methodologists.

Two competitors deserve a real defense plan. The first is a conversational-research vendor launching an enterprise employee-intelligence SKU with API distribution. The defense is to move first on a brand-partner API in beauty retail and lock the proof case before they rotate. The second is a knowledge marketplace player hiring a vertical head in retail or restaurants and shipping a multi-tenant graph. The defense is to out-vertical and out-domain them (strategy versus operations) before their next funding round.

The full competitive analysis (each adjacent category, the specific rotator, the rotation signal, the defensive move) is on the category map.

Early commercial deployments are running in beauty retail and in restaurant networks. The first proof cases validated the unit economics at three cents per enriched conversation and the protocol fidelity at the ten typed dimensions per conversation. The next two quarters expand into the brand-partner API and the AI-agent feed on those same deployments.

If your vertical is one of the ones described above and you want to discuss what a deployment shape would look like in your context, the contact form is the right starting point.