FIRSTVAL

01Marketing Twin

Spend that proves itself.

A live model of demand generation — channels, audiences, creative, and spend — simulated against pipeline and revenue before the budget is committed.

North Star

Cost per qualified pipeline dollar

ValueNorth StarKPI / driver treeCausal modelDigital twinSimulationOptimizationInterventionAgentic executionReal-world outcomeAttributionEconomic value createdLearning

The stream

Automation → Capability → Value driver → Financial value

Automation

Always-on attribution and spend reallocation

Capability

Know the marginal return of every channel weekly

Value driver

Marketing efficiency

Financial value

Lower CAC, higher pipeline per dollar

Automation

Agentic brief, copy and variant generation

Capability

Ship 5× more tested creative at same headcount

Value driver

Speed to market

Financial value

Earlier revenue recognition per campaign

Automation

Audience and offer matching from the CDP

Capability

Right offer, right segment, governed consent

Value driver

Conversion quality

Financial value

Higher win rate on sourced pipeline

Value system

Value ≠ North Star ≠ KPI ≠ driver ≠ process metric.

The translation layer that connects a corporate objective to something an operating team can actually move on Monday.

Value

Profitable demand

Economic outcome

Contribution margin on sourced revenue

North Star

Cost per qualified pipeline dollar

Outcome KPIs

Sourced pipelinePipeline-to-revenue conversionCAC payback

Driver metrics

Channel marginal returnCreative win rateSegment responseLead quality score

Process metrics

Brief-to-live daysVariant test countConsent coverageData freshness

Causal model

Budget allocationChannel exposure by segmentQualified response ratePipeline createdWin rate on sourced dealsContribution marginEnterprise value

Value leakage

Where the value goes missing today.

Leakage is multiplicative. Every gate that stays leaky discounts everything built upstream of it.

Targeting

Spend against segments that never convert.

Wasted media

Hand-off

Qualified leads aged out before first touch.

Lost pipeline

Attribution

Last-click credit hides the real driver.

Misallocated budget

Consent

Audiences rebuilt after consent failures.

Rework and exposure

Digital twin

Simulate the interventions before anyone funds them.

Instead of implementing ten recommendations and learning the result in six months, the twin stacks them first.

Cost per qualified pipeline dollar

Current$0.42
Cut zero-uplift channels$0.37
Shift budget to proven segments$0.33
5× creative variant testing$0.30
Lead routing inside 5 minutes$0.28
Combined, optimized$0.26

Guardrails

  • Brand safety score maintained
  • No special-category inference
  • Share of voice not below floor

Constraints

  • Quarterly budget cap
  • Agency contract commitments
  • Regional consent regimes

Automation

What runs without asking.

Budget reallocation loop

Weekly mix optimization against pipeline value, not clicks.

Attribution Agent

Creative production line

Brief → draft → brand check → variant set, human approved.

Content Ops Agent

Campaign QA + consent gate

Blocks launch when consent basis or claims fail policy.

Guardrail Agent

Campaign Planner Agent

Plans spend and sequencing against a pipeline target.

LLM reasoning + optimization solver

Attribution Agent

Maintains multi-touch and uplift models, explains shifts.

SLM classification + causal uplift

Content Ops Agent

Generates and routes creative through brand and legal checks.

LLM + RAG on brand corpus

Integrations

Where it plugs into the domain.

Vendor-agnostic by design. The twin reads and writes through whatever stack the domain already runs on.

Systems of record

  • CRM (Salesforce, HubSpot, Dynamics)
  • Marketing automation
  • CDP

Channels

  • Ad platform APIs
  • Email / SMS
  • Web & product analytics

Data & runtime

  • Warehouse / lakehouse
  • Metric layer
  • MCP connectors
  • Model gateway

Domain platforms we connect

CRM

HubSpotSalesforce Sales CloudMicrosoft Dynamics 365Zoho CRMPipedrive

Marketing automation

Marketo EngageHubSpot Marketing HubSalesforce Marketing CloudBrazeKlaviyoIterable

CDP & audiences

SegmentTealiumAdobe Experience PlatformmParticleSalesforce Data Cloud

Advertising & channels

Google AdsMeta AdsLinkedIn AdsThe Trade DeskAmazon Ads

Analytics & content

GA4AmplitudeAdobe AnalyticsContentfulAdobe Experience ManagerFigma

Data & runtime

SnowflakeDatabricksBigQuerydbtFivetranMCP connectors

Plus anything else with an API — connectors are added per engagement, not sold as a platform lock-in.

Value

The levers, and what they move.

Cost

Kill spend on channels with no marginal pipeline effect.

CAC · Cost per qualified pipeline dollar

Growth

Shift budget to segments with proven uplift.

Sourced pipeline · Conversion rate

Speed

Compress brief-to-live cycle time.

Days from brief to campaign live

Risk

What could go wrong, and what stops it.

Every risk in this domain has a named containment in the runtime — not a slide.

Generated claims that cannot be substantiated

Claims policy engine + human approval before publish

High

Profiling without a lawful basis

Consent state enforced at audience build time

High

Attribution model drift after channel change

Holdout tests and uplift re-estimation each cycle

Medium

Compliance · Security

Built into the runtime, not bolted on.

EU AI Act

Limited risk — transparency obligations (generated content, personalization).

  • Disclose AI-generated or AI-assisted content where required
  • Log model, prompt and data lineage for every published asset
  • Human review before any external publication

GDPR

  • Lawful basis and consent state enforced at audience build time
  • No special-category inference for targeting
  • Data minimization: hashed identifiers only in the model layer
  • DSAR and erasure propagated to the twin and vector store

Security

  • SSO/SCIM with least privilege
  • Tenant-isolated retrieval indexes
  • Prompt-injection filtering on ingested web content

Controls & oversight

  • Brand and claims policy engine
  • Spend approval thresholds
  • Full audit trail of agent actions

Agentic execution

Agents earn authority. They are not given it.

No agent in this twin controls anything it has not first proven in replay, evaluation, simulation and shadow.

01

Historical replay

Re-run the last 12 months. Would the agent have been right?

02

Offline evaluation

Scored against held-out outcomes, not opinion.

03

Digital twin

Simulated against the causal model under stress.

04

Shadow mode

Runs live, decides nothing. Divergence is logged.

05

Human recommendation

Proposes; a person executes and rates it.

06

Bounded pilot

One segment, capped exposure, hard rollback.

07

Human-supervised execution

Acts inside thresholds, humans approve exceptions.

08

Progressive autonomy

Authority widens only where evidence widened.

The unit of value

Don't buy transformation. Buy measurable movement.

This twin is contracted the way it is engineered: a baseline, a target, a guardrail, and an attribution method agreed in advance.

Baseline

Cost per qualified pipeline dollar $0.42

Target

$0.30 within two quarters

Guardrail

Sourced pipeline volume never below baseline

Proof / attribution

Geo and audience holdouts, reconciled to CRM revenue