Sales Motion / Cognitive Model Architecture

Sales Motion
Cognitive Model Architecture

An architecture for controlling how understanding progresses across the Sales Motion — from foresight and company evidence to buying insight, customer narrative and formal response — through modular, evidence-gated pipelines.

01

Cognitive Model: defines what must be understood next, using what evidence, and under what reasoning boundaries.

02

Evidence-Gated Pipelines: govern how each cognitive step is executed, validated and converted into an approved artifact.

03

Execution Components: models, agents, APIs, databases, search and humans perform the work inside the architecture — they do not define it.

00 / Why This Matters

The point is not better orchestration.
The point is better business outcomes.

Sales Motion is designed to move the organisation upstream: from reacting to already-defined demand toward identifying, validating and shaping evidence-supported customer needs earlier — while improving the speed and quality of the response once demand becomes explicit.

−70%
Request-to-Offer time
Real case outcome from the combined operating model.
+50%
Winning ratio
Real case outcome from the combined operating model.
+35%
NPS
Real case outcome from the combined operating model.
<2 days
Strategic Validation Loop
Real case outcome from the combined operating model.
90%
Utilisation
Real case outcome from the combined operating model.
+20%
Net profit
Real case outcome from the combined operating model.
Evidence note: these figures are presented in the Sales Motion material as outcomes of the combined operating model — demand creation, pull-based capacity planning, Service Areas, the Impact Opportunity Platform and RFP Orchestrator — not as a claim that AI, software or reorganisation alone produced them.

What the Sales Motion can systematically govern and improve

The architecture should be judged through measurable operating and business outcomes — not by how sophisticated the orchestration looks.

Signal → Validation time Signal-triggered opportunity ratio Proposal accuracy Request-to-Offer time Winning ratio NPS Utilisation Sales cost Customer churn Net profit
If these are the outcomes we want, the architectural question becomes: what exactly should we orchestrate?

01 / Opening Thesis

The distinctive idea of Sales Motion is not the existence of modules, databases, pipelines or even agents. Those are implementation components. The architecture is really about controlling how cognition progresses from one state of understanding to the next.

Implementation-first architecture

Starts from agents, models, tools, APIs and orchestration. The central question becomes: who acts next?

Cognitive Model Architecture

Starts from the required progression of understanding. The central question becomes: what must become known next, from what evidence, and under what conditions may the system proceed?

02 / Core Architectural Principle

Do not orchestrate your agents. Orchestrate your Cognitive Model.

The architecture should first define the sequence of cognitive states, evidence requirements, reasoning boundaries, validation gates and artifact contracts. Only then should it decide which model, agent, API, deterministic service, database query or human performs each operation.

01Cognitive ModelWhat must happen intellectually?
What must be understood next?What evidence is required?What inference is allowed?What must be validated?What artifact represents the new state?
02Evidence-Gated PipelineUnder what rules may the cognition happen?
Acquire evidenceNormalize / classifyCheck provenance + freshnessMeasure sufficiencyReasonValidatePublish governed artifact
03Execution ComponentsWhat performs the work?
LLMsAgentsDeterministic codeAPIsSearchDatabasesKnowledge basesHuman validation
01Sales MotionThe business system
Defines the end-to-end progression from sensing to customer action
02Cognitive Model ArchitectureThe orchestration logic
Cognitive statesEvidence requirementsState transitionsArtifact contractsValidation gates
03Modules & CapabilitiesWhat executes parts of the motion
The ScannerForesight CreationBuying InsightBuying NarrativeRFP OrchestratorSimulator
04Execution TechnologyImplementation resources
AgentsLLMsAPIsSearchDatabasesKnowledge basesDeterministic codeHuman validation

03 / System Flow

The cognitive progression from external change to customer action

The Sales Motion is not one monolithic workflow and not a chain of agents. It is a sequence of cognitive states implemented through modular pipelines. Each module can operate independently, while governed artifacts connect the modules horizontally when the evidence is sufficient and the next cognitive step is justified.

01 / Sense

Foresight Creation

Detect and structure emerging change, weak signals, regulation, technology and market developments.

Output: validated foresight / signal
02A / Establish reality

Company Summary

Build an evidence-backed current-state baseline of the company, stakeholders, strategy and relevant developments.

Output: governed company context
02B / Establish pressure

Market Intelligence

Connect the company to market, competitive, industry, regulatory and structural pressures.

Output: market pressure analysis
03 / Form hypothesis

Buying Insight & Hypothesis

Combine foresight, company reality and market pressure into a testable customer-specific hypothesis.

Output: buying insight
04 / Frame dialogue

Buying Narrative

Turn analytical findings into a coherent executive story and a customer decision conversation.

Output: buying narrative
05 / Formalize response

RFP Orchestrator

Map formal requirements to capabilities, approved evidence, gaps, response structure and human validation.

Output: validated proposal / RFP package

04 / Architecture Stack

Cognition first. Execution components last.

The architecture separates the cognitive model from pipeline mechanics and execution technology. The Sales Motion therefore remains independent of any single module or tool: The Scanner, RFP Orchestrator, models, agents, databases and other capabilities can evolve without changing the required cognitive progression.

01Cognitive Model LayerWhat must become known
Signal interpretationCurrent-state synthesisPressure analysisHypothesis formationNarrative constructionRequirement-response reasoning
02Evidence-Gated Pipeline LayerHow cognition is controlled
AcquireNormalizeClassifyCheck provenanceMeasure coverageReasonValidatePublish artifact
03Execution ComponentsWhat performs the work
LLMsAgentsDeterministic servicesAPIsSearchDatabase queriesHuman validation
04Shared Data & Knowledge LayerWhat modules can consume
Foresight KBCompany / Account DataMarket Intelligence KBService & Capability KBReference / Case KBProposal / RFP KBCRMCoreSignal / VainuService CatalogResource Planning
05Governance & State LayerWhat makes progression defensible
ProvenanceFreshnessEvidence coverageValidation stateRun stateAudit trailHuman approval
06Governed Outcome ArtifactsWhat humans and downstream modules consume
ForesightCompany SummaryMarket IntelligenceBuying InsightBuying NarrativeProposal / RFP Response

05 / Cognitive Layer

The Cognitive Model is the orchestration logic

The system advances by changing the state of understanding, not by merely handing work from one agent to another. Each module therefore owns a distinct cognitive question, evidence requirement, reasoning boundary and stopping criterion.

Foresight Creation

What is changing — and why could it matter?

Detects, classifies and validates emerging signals before they become customer-specific.

Deterministic: provenance, taxonomy, freshnessGenAI: synthesis, interpretation

Company Summary

What is currently true about this company?

Builds a defensible current-state baseline from company, account and stakeholder evidence.

Deterministic: source status, dates, entity checksGenAI: synthesis of fragmented evidence

Market Intelligence

What external pressures surround this company?

Connects account reality to market, regulation, competition and structural change.

Deterministic: source mapping, signal linkageGenAI: pressure analysis

Buying Insight & Hypothesis

Why could this matter to this customer now?

Separates observed facts from inference, forms a hypothesis and identifies what still requires validation.

Deterministic: evidence gates, missing-evidence flagsGenAI: hypothesis formation

Buying Narrative

What is the coherent story worth discussing?

Turns analysis into a customer-relevant executive narrative without converting hypotheses into facts.

Deterministic: source boundaries, claim rulesGenAI: narrative construction

RFP Orchestrator

How do we answer the recognized need defensibly?

Maps formal requirements to capabilities, evidence, gaps and approved response components.

Deterministic: requirement coverage, completeness, approvalsGenAI: drafting, response synthesis

06 / Vertical × Horizontal

Vertical cognition. Horizontal progression.

Vertically, each cognitive step is governed end to end. Horizontally, the system advances only when an approved artifact satisfies the contract for the next step. This makes progression a property of the Cognitive Model Architecture rather than of agent choreography.

Vertical: evidence-gated module

Inputs / Authorized Sources
Evidence Acquisition
Normalization / Classification / Security
Provenance + Freshness Checks
Evidence Coverage / Sufficiency
Governed Cognitive Analysis
Deterministic Validation Gates
Approved Artifact

Horizontal: artifact-to-artifact contracts

Validated Foresight
Market Intelligence
Company Summary
Buying Insight
Market Pressure Analysis
Buying Insight
Buying Insight
Buying Narrative
Validated Customer Need
RFP Orchestrator
Proposal / RFP Outcome
CRM + Learning Loop

07 / Where The Scanner Fits

The Scanner is a capability inside the Sales Motion

The broader architecture is the Sales Motion. The Scanner and its modules contribute selected sensing, company-context, market-intelligence and buying-insight capabilities inside that motion. Other components — such as Buying Narrative, Simulator and RFP Orchestrator — continue the cognitive progression.

01Sales MotionEnd-to-end cognitive progression
ForesightCompany RealityMarket PressureBuying HypothesisBuying NarrativeProposal / RFP Response
02The ScannerSelected sensing & analysis capabilities
Foresight-driven sensingCompany SummaryMarket IntelligenceBuying Insight & Hypothesis
03Other Motion ComponentsContinue the flow
Buying NarrativeSimulatorProposal CreationRFP OrchestratorHuman / Team Validation

08 / Shared Data & Knowledge Layer

Shared resources, not duplicated databases

The Cognitive Model Architecture uses shared governed sources and knowledge services without collapsing all modules into one reasoning process. Common evidence and metadata create consistency; module-specific cognitive logic creates precision.

Foresight Knowledge

  • Signal taxonomy
  • Source register
  • Trend / weak-signal objects
  • Regulatory and technology change

Company / Account Data

  • CRM
  • CoreSignal
  • Vainu / external company data
  • Company website and public evidence

Market Intelligence

  • Industry structure
  • Competitive evidence
  • Market pressure taxonomy
  • Regulatory context

Service & Capability Knowledge

  • Service catalog
  • Capabilities
  • Delivery constraints
  • Resource / capacity data

Reference / Case Knowledge

  • Customer cases
  • Outcome evidence
  • Approved proof points
  • Reusable references

Proposal / RFP Knowledge

  • Requirement taxonomy
  • Approved claims
  • Proposal components
  • Win / loss learning

09 / Buying Narrative

The bridge from analysis to customer dialogue

Buying Narrative is a distinct cognitive state between analytical hypothesis and formal response. It does not simply reformat Buying Insight: it reframes validated analysis into a decision-oriented customer conversation while preserving the boundary between evidence, interpretation and open questions.

Buying Insight

What could be happening, why could it matter, and what must be validated?

Analytical artifact. Evidence and inference remain explicitly separated.

Buying Narrative

What is the coherent customer story and leading idea?

Conversation artifact. It frames the situation, evidence, implications, unresolved questions and proposed executive dialogue.

RFP / Proposal

How do we respond once the need is recognized or formalized?

Response artifact. Requirements, capability evidence, claims, gaps and approvals become explicit.

10 / Artifact Contracts

What each module should hand forward

The horizontal architecture becomes reliable when a cognitive state is represented by a governed artifact rather than a loose text handoff. Each artifact therefore carries known fields, evidence state, unresolved uncertainty and rules for downstream consumption.

ModuleCore artifactMinimum contract ideaTypical consumer
Foresight CreationValidated Foresight / SignalSignal, source, date, taxonomy, relevance, confidence, provenanceMarket Intelligence, Buying Insight
Company SummaryGoverned Company ContextCurrent facts, stakeholders, events, source dates, confidence, unresolved gapsBuying Insight, Sales
Market IntelligenceMarket Pressure AnalysisPressure, evidence, affected capability / business area, direction, strength, uncertaintyBuying Insight
Buying InsightBuying Insight & HypothesisObserved evidence, hypothesis, causal logic, uncertainty, validation questions, opportunity directionBuying Narrative, Sales
Buying NarrativeCustomer NarrativeLeading idea, evidence-backed story, implications, decision questions, proposed working sessionCustomer dialogue, Proposal
RFP OrchestratorValidated Response PackageRequirements, mapped evidence, response text, gaps, completeness state, human approvalsCustomer / procurement process

11 / Design Principles

What makes the system coherent

Cognition before orchestrationDefine the cognitive sequence, evidence and gates first. Select agents, models and tools only after the reasoning architecture is explicit.
Evidence before inferenceObserved evidence, derived analysis and hypotheses remain distinguishable throughout the flow.
Explicit state transitionsHorizontal connections are governed artifact contracts that authorize the next cognitive step, not informal agent-to-agent handoffs.
Human authorityCustomer-facing decisions, narrative ownership and formal publication remain subject to human validation.
Shared knowledgeCommon knowledge bases create consistency while allowing module-specific cognitive models.
Measured sufficiencyCoverage, provenance, freshness and unresolved gaps determine whether analysis may proceed.
Reusable learningCRM outcomes, proposal results and win/loss learning feed the knowledge layer instead of disappearing into local documents.
Upstream orientationThe system aims to create a reason to engage before demand is fully formalized, while remaining evidence-backed.