Skip to content
RAY BEHARRY

AI Marketing Department · Workflows

Fourteen workflows. One operating rhythm.

Each workflow is a machine-readable spec with a trigger, agent dependencies, steps, data contracts, human approval gates, metrics and a feedback loop.

Walkthrough

Workflow walkthrough · 14 shipped workflowsStep-through of the real workflow spec, not a live run
Choose a workflow

Market signals → synthesis → strategic implications → recommended actions. A weekly, sourced brief that reduces market noise to at most three implications, each with an owner and the decision it should change.

Ask it: “Run the market pulse” · or run /wf-market-pulse

Step 1 / 7

Listen · Agent

Collect and log signals

Market Research Analyst

/market-pulse

What it does

Gather new signals from market-signals plus public sources; log each with source URL, date and evidence class. Reject anything unsourced.

Reads

Company Context · ICP Library · Market Signals

Writes back to

Market Signals

Produces: signal-log

Each step is taken from the workflow’s machine-readable spec in the v1.1.0 package. A step only completes when its output passes its contract and proof checks.

Reference

Every workflow, step by step.

01 · /wf-market-pulse

Market Pulse

Market signals → synthesis → strategic implications → recommended actions. A weekly, sourced brief that reduces market noise to at most three implications, each with an owner and the decision it should change.

Book frameworks: Operating Loop

  1. 1Collect and log signalsMarket Research Analyst
  2. 2Check customer language shiftsVoice of Customer Analyst
  3. 3Verify competitor signalsCompetitor Monitoring Agent
  4. 4Synthesize patterns and implicationsMarket Research Analyst
  5. 5Structural trend checkMarket Trend Analyst
  6. 6Evidence and quality checkMarketing QA Agent
  7. 7Approve implications and assign ownersAI Marketing Chief of Staff· approval: Head of Product Marketing (or delegated PMM)

02 · /wf-positioning-engine

Positioning Engine

ICP → Category → Problem → Differentiation → Proof → Positioning → Validation. Builds a stage-aware positioning canvas from evidence, tests it with a synthetic audience as a hypothesis plus a real-world subhead test plan, and only enters positioning-canon after a named human approves it.

Book frameworks: Adaptive Positioning, Message Supply Chain

  1. 1Confirm the target ICPICP Research Analyst
  2. 2Choose the category frameCategory Strategy Architect
  3. 3Define the problem in the buyer's wordsVoice of Customer Analyst
  4. 4Map differentiation against alternativesDifferentiation Analyst
  5. 5Attach proof to every claimValue Proposition Architect
  6. 6Draft the positioning canvasPositioning Engine
  7. 7Validate with synthetic audience and subhead test planPositioning Engine
  8. 8Brand taste gateBrand Strategy Guardian
  9. 9Approve into positioning canonAI Marketing Chief of Staff· approval: Head of Product Marketing (with CMO sign-off for category changes)

03 · /wf-messaging-engine

Messaging Engine

Approved positioning → message kernels → persona variants → channel variants → quality review → approval into message-memory. Turns canon positioning into a reusable kernel set and variant grid that passes a Truth Pass, a Taste Gate and a proof check before any line is approved for use.

Book frameworks: Message Supply Chain

  1. 1Load approved positioningMessaging Strategist
  2. 2Write message kernelsMessage Kernel Architect
  3. 3Generate persona variantsMessaging Strategist
  4. 4Generate channel variantsContent Repurposing Agent
  5. 5Draft sample assetsContent Production Agent
  6. 6Truth Pass and decision-stage proof checkMarketing QA Agent
  7. 7Taste GateBrand Strategy Guardian
  8. 8Approve into message memoryAI Marketing Chief of Staff· approval: Head of Product Marketing (brand lead for voice exceptions)

04 · /wf-learning-launch

Learning Launch

Narrative hypotheses → launch plan → experimentation → Day 7/14/30/60 reviews → learning memory. Treats a launch as a structured experiment: at most three narrative bets, a staggered Day −30 → +60 plan, tagged experiments, scheduled reviews that read, tune, consolidate and extend, and a Launch Learnings Memo that updates canon only with approval.

Book frameworks: Impact Chain, Learning Launch

  1. 1Define narrative hypothesesLaunch Strategy Architect
  2. 2Build the Day −30 → +60 launch planGTM Planning Agent· approval: Launch owner (PMM lead) with budget holder for any spend
  3. 3Channel activation and field talk tracksChannel Activation Planner
  4. 4Field talk tracks per betSales Talk Track Agent
  5. 5Design launch experimentsLaunch Experiment Designer
  6. 6Day 7 readLaunch Learning Analyst
  7. 7Day 14 tuneLaunch Learning Analyst
  8. 8Day 30 consolidateLaunch Learning Analyst
  9. 9Day 60 extend or evolve and write learning memoryLaunch Learning Analyst· approval: Head of Product Marketing

05 · /wf-sales-enablement

Sales Enablement Engine

Buyer question → ESOT retrieval → LPNS response → seller feedback → updated guidance. Answers live field questions from the Enablement Source of Truth in a Listen-Prove-Next-Step format, captures seller feedback, runs the objection lab, and updates ESOT only after approval — with pricing and roadmap answers confirmed by their owners.

Book frameworks: LPNS Pattern, Enablement Context Loop

  1. 1Classify the buyer questionEnablement Context Architect
  2. 2Retrieve from ESOTEnablement Context Architect
  3. 3Draft LPNS answerSales Talk Track Agent
  4. 4Owner confirmation for pricing and roadmapPricing & Packaging Analyst· approval: Pricing owner (pricing) or Product lead (roadmap); PMM for other topics
  5. 5Capture seller feedbackField Feedback Analyst
  6. 6Run the objection labObjection Handling Agent
  7. 7Update ESOT guidanceAI Marketing Chief of Staff· approval: Sales Enablement lead with PMM owner

06 · /wf-competitive-response

Competitive Response

Competitor signal → verification → strategic interpretation → counter messaging → battlecard update. Verifies a competitor move against sources and baseline, interprets it with confidence and a falsifier, ships a 72-hour Counter Kit built on the Four Cs, and refreshes battlecards after human approval.

Book frameworks: CI Loop

  1. 1Capture the competitor signalCompetitor Monitoring Agent
  2. 2Verify the signalCompetitor Monitoring Agent
  3. 3Language and pricing forensicsLanguage Drift Analyst
  4. 4Pricing and packaging checkPricing & Packaging Analyst
  5. 5Strategic interpretationCompetitive Strategy Analyst
  6. 6Build the 72-hour Counter KitCompetitive Response Agent· approval: Head of Product Marketing (legal review for comparative claims)
  7. 7Update the battlecardBattlecard Builder· approval: Sales Enablement lead

07 · /wf-seo-growth

SEO Growth Engine

Keyword research → search intent → content brief → SEO optimization → search performance review. Finds keyword opportunities that match the ICP and positioning, using volumes only from Semrush, Ahrefs or Search Console exports, briefs and optimizes content, and reviews measured search performance to decide what to refresh next.

Book frameworks: Operating Loop

  1. 1Keyword opportunity researchKeyword Research Agent
  2. 2Classify search intentSearch Intent Analyst
  3. 3Write the SEO content briefContent Strategy Agent
  4. 4Technical SEO checkTechnical SEO Auditor
  5. 5On-page optimizationOn-Page SEO Optimizer· approval: Content lead (web owner publishes)
  6. 6Search performance reviewSearch Visibility Analyst
  7. 7Prioritize refresh and next clustersAI Marketing Chief of Staff· approval: Head of Content / SEO lead

08 · /wf-demand-generation

Demand Generation Engine

Audience → campaign hypothesis → offer → channel strategy → content → measurement. Builds campaigns as testable hypotheses for a defined ICP audience, with a proof-backed offer, a distribution plan, drafted content and pre-agreed measures; spend and publishing happen only after human approval.

Book frameworks: Operating Loop, Impact Chain

  1. 1Define the audienceAudience Growth Analyst
  2. 2Write the campaign hypothesisCampaign Strategy Agent
  3. 3Design the offerLead Magnet Architect
  4. 4Channel and distribution planDistribution Planner· approval: Demand generation lead with budget holder
  5. 5LinkedIn content planLinkedIn Strategy Agent
  6. 6Draft campaign contentContent Production Agent· approval: Content lead (channel owners publish)
  7. 7Measure against the hypothesisGTM Performance Analyst

09 · /wf-lifecycle

Lifecycle Engine

Audience segment → lifecycle stage → email sequence → performance analysis → optimization. Maps a segment to its lifecycle stage, drafts the sequence from approved messaging, and optimizes from measured ESP results. Agents draft; humans approve and send from the ESP.

Book frameworks: Operating Loop, Impact Chain

  1. 1Define the audience segmentLifecycle Marketing Agent
  2. 2Map the lifecycle stageLifecycle Marketing Agent
  3. 3Design the sequenceNurture Sequence Architect
  4. 4Draft email copyEmail Copywriter
  5. 5Strategy and send approvalEmail Strategy Agent· approval: Lifecycle/Email owner (sends from the ESP)
  6. 6Analyze sequence performanceEmail Performance Optimizer
  7. 7Recommend optimizationsEmail Performance Optimizer

10 · /wf-conversion

Conversion Engine

Funnel diagnosis → hypothesis → landing page variants → experiment → validated learning. Finds the biggest measured leak in the funnel, writes a falsifiable hypothesis, drafts variants, runs a pre-registered experiment with stop and promotion rules, and records the validated learning.

Book frameworks: Operating Loop, Impact Chain

  1. 1Diagnose the funnelFunnel Analysis Agent
  2. 2Audit the landing pageLanding Page Strategist
  3. 3Draft landing page variantsConversion Copy Agent
  4. 4Pre-register the experimentA/B Experiment Designer· approval: Growth/CRO lead (web owner launches the test)
  5. 5Read the experimentCRO Performance Analyst
  6. 6Record validated learningAI Marketing Chief of Staff· approval: Growth/CRO lead

11 · /wf-gtm-performance

GTM Performance Engine

Campaign results → Narrative Fit → Engagement Quality → Sales Momentum → recommendations. Reads measured results through the book's three lenses, links them to revenue outcomes on a Command Board, and recommends Stop / Start / Scale moves that a human approves before any budget or plan changes.

Book frameworks: Impact Chain

  1. 1Collect campaign resultsGTM Performance Analyst
  2. 2Narrative FitNarrative Fit Analyst
  3. 3Engagement QualityEngagement Quality Analyst
  4. 4Sales MomentumSales Momentum Analyst
  5. 5Build the Command BoardGTM Performance Analyst
  6. 6Learning loop reviewLearning Loop Optimizer
  7. 7Approve Stop / Start / ScaleAI Marketing Chief of Staff· approval: CMO or VP Marketing (budget holder for Scale moves)

12 · /wf-executive-brief

Executive Marketing Brief

Market intelligence + campaign performance + competitive changes + field feedback → prioritized executive decisions. Condenses the system's latest outputs into one brief whose decision slide reads What moved → Why → What we'll change now, approved by an executive and logged to decision-history.

Book frameworks: Operating Loop, Impact Chain

  1. 1Market intelligence summaryMarket Research Analyst
  2. 2Campaign performance summaryGTM Performance Analyst
  3. 3Competitive changesCompetitive Strategy Analyst
  4. 4Field feedbackField Feedback Analyst
  5. 5Draft the brief and decision slideAI Marketing Chief of Staff
  6. 6Evidence QAMarketing QA Agent
  7. 7Executive approval and decision logAI Marketing Chief of Staff· approval: CMO (or executive sponsor)

13 · /wf-social-community

Social & Community Engine

Approved kernels → organic platform mix → tested hooks → tagged native social posts → founder and executive thought leadership → community conversation plan → human approval before anything goes live → audience-quality reads → logged learning. Organic social is judged by ICP share of engagers and qualified conversations, never by follower counts.

Book frameworks: Impact Chain, Message Supply Chain

  1. 1Platform mix and pillars from approved narrativeSocial Strategy Agent
  2. 2Tested hook variantsHook Creator
  3. 3Native posts tagged Persona × Stage × ChannelSocial Post Writer
  4. 4Executive and founder thought leadershipThought Leadership Agent
  5. 5Community engagement planCommunity Manager
  6. 6Pre-publish QA and human approvalMarketing QA Agent· approval: Social lead plus each named publisher (executives approve their own posts)
  7. 7Audience-quality readsEngagement Quality Analyst
  8. 8Log learning and Stop / Start / ScaleSocial Strategy Agent

14 · /wf-editorial

Editorial Engine

Content gap analysis → offer and angle → brief → long-form draft → edit (Truth Pass, Taste Gate, proof check) → human approval before publishing → newsletter and repurposing → performance review on engaged ICP readers and assisted pipeline → learning logged. Builds an editorial calendar from unanswered buyer questions and ships only pieces that are sourced, proven and approved.

Book frameworks: Impact Chain, Message Supply Chain

  1. 1Content gap analysisContent Gap Analyst
  2. 2Define offer and angleOffer Strategist
  3. 3Write the editorial briefContent Strategy Agent
  4. 4Draft the long-form pieceLongform Writer
  5. 5Edit, Truth Pass, Taste Gate and proof checkContent Editor
  6. 6Human approval before publishingAI Marketing Chief of Staff· approval: Content lead (offer owner for any commercial term)
  7. 7Build the newsletter issueNewsletter Editor
  8. 8Performance review and learningEngagement Quality Analyst