jurjen-gtm-engineer/gtmskills

55 skill entries

• ai-use-case-gate

  Run a proposed AI use case for a GTM team through a strict pre-build gate; constraint check first, then the chocolate teapot ship test, then capability-ladder placement, guardrails, and a one-page go/no-go verdict.

  Self-containedInstructions only2,702 words
• b2b-or-b2c

  Classify whether a company's business model is B2B, B2C, or both

  Self-containedInstructions only359 words
• battlecard-builder

  Build competitive battlecards in 30 minutes using AI-powered research and a 3-type framework (Core Competitor, Quick Reference, Objection Handler). Battlecards that are simple, current, and actually get used mid-call.

  Self-containedInstructions only861 words
• blueprint-swarm

  Multi-agent call-data analysis at scale. Reads hundreds of call transcripts and CRM exports in parallel, extracts churn timelines, win patterns, competitive intel, product gaps, and playbook material, with source-tagged quotes and an Opus auditor that kills hallucinated output. Based on Jordan Crawford's open-source Blueprint Swarm.

  Self-containedInstructions only1,435 words
• bowtie-benchmark

  Interview the user for their bowtie revenue metrics, benchmark each against companies at their ACV band, score ahead/near/behind, and write a scorecard report with per-metric leak diagnosis.

  Reaches the webInstructions only1,389 words
• claygent-builder

  Generate production-ready Claygent configurations (system prompt + task prompt + JSON schema + examples), test them via Clay webhook, and iterate until quality hits 8.0+

  Self-containedActs undeclared1,813 words
• claygent-prompt-generator

  Generate cache-optimized Claygent prompts with all static logic first and {{variables}} at the bottom for Clay API cached inputs

  Self-containedInstructions only1,419 words
• clean-job-titles

  Standardize verbose LinkedIn job titles into clean, usable formats

  Self-containedInstructions only416 words
• company-goals

  Infer company strategic goals from their job postings and hiring patterns

  Self-containedInstructions only458 words
• company-mission

  Extract a company's mission statement from their LinkedIn or website

  Self-containedInstructions only248 words
• compound-growth-check

  Diagnose whether a company's growth actually compounds. Takes 6+ quarters of ARR, computes the first and second derivatives, classifies the trajectory (compounding, inflection, decompounding, decay), places the company on the 10-state growth ladder, and writes a one-page verdict.

  Self-containedActs undeclared1,358 words
• context-gap-analysis

  Enumerate required context, check what exists, and find the simplest path forward before any task. Prevents hallucinations by forcing the agent to verify assumptions before acting.

  Self-containedActs undeclared1,430 words
• conversational-intelligence

  Extract structured intelligence from call transcripts, from any transcript source (Fireflies, Gong, or plain files exported locally). Mines conversations for handoff context, competitive intelligence, expansion signals, and closed-lost analysis.

  Self-containedInstructions only1,623 words
• crm-enrichment-cost-estimate

  Triggers when a user wants to estimate yearly Clay data credits and actions for enriching records they already have in a CRM, with phrasings like "how many Clay credits will CRM enrichment use", "scope credits and actions", "what does enriching 20,000 contacts cost per year", or "size the Clay plan for this use case". Owns the audience size, credits and actions per enrichment field, signals and exports, and the yearly totals per record. Follows the model of Clay's public Data Credit and Actions Scoping Template. Not for pricing a new market sourced from scratch (tam-data-cost-estimate) and not for auditing what a live workspace already spends.

  Reaches the webActs undeclared679 words
• crm-scorecard

  Turn raw CRM CSV exports (HubSpot, Salesforce, Attio, Pipedrive, or any CRM) into a bowtie conversion scorecard, computing CR1-CR8 where the data allows, benchmarking against ACV-band tables, and naming the top two revenue leaks.

  Self-containedInstructions only2,665 words
• customer-dossier

  Build a ground-truth customer dossier per account from CRM exports, call transcripts, billing, product usage, and support data. One sorted timeline, provenance on every field, conflicts surfaced instead of hidden. The dossier describes, never reasons.

  Self-containedInstructions only1,015 words
• customer-evidence-first

  Start GTM strategy from why customers actually bought, not what you sell (Jordan Crawford)

  Reaches the webInstructions only860 words
• data-point-research

  Design custom data points that predict fit but can't be bought from standard providers

  Self-containedInstructions only888 words
• email-opening-line

  Generate personalized email opening lines that hook with relevance

  Self-containedInstructions only621 words
• email-subject-line

  Generate personalized email subject lines under 8 words using prospect research

  Self-containedInstructions only568 words
• epistemic-context-grounding

  Ground implementation decisions in domain knowledge before designing solutions. Prevents over-engineering by checking what documentation exists, making assumptions explicit, and verifying them against canonical sources. Core principle - know what you don't know before designing.

  Self-containedActs undeclared2,081 words
• follow-up-sequences

  Email sequence strategy for multi-touch campaigns with value prop rotation

  Self-containedInstructions only1,968 words
• free-first-domain-resolver

  Resolve a list of company names to verified domains for a fraction of a cent per name instead of a per-row vendor fee. Runs a cost waterfall (owned Google Maps table and Google Knowledge Graph for free, then one cheap search, then Google Places) and verifies every domain against what the page declares about itself (title, Open Graph, JSON-LD), never the SSL certificate. Refuses to guess: anything it cannot confirm is flagged \"needs review\" instead of returning a confident wrong domain. Use when someone has a CSV of company names and needs domains (the same move works for LinkedIn URLs and parent companies). Based on Jordan Crawford's free-first resolver idea (Blueprint GTM). Not for finding contact emails or enriching people.

  Needs a keyActs undeclared997 words
• glassdoor-rating

  Find and interpret a company's Glassdoor rating for culture insights

  Self-containedInstructions only517 words
• gtm-ai-brief

  Turn a GTM task someone keeps doing manually (or keeps prompting ad hoc) into a written, testable AI brief using the Trigger / Inputs / Steps / Output / Guardrails scaffold, tested against synthetic data before it ships.

  Self-containedInstructions only2,069 words
• gtm-diagnostic

  5-week GTM diagnostic sprint that stacks four frameworks (Revenue Architecture, Bowtie Analytics, Growth Architecture, Insight Engineering) to measure the full revenue engine, name the dominant cause, forward-project the probability of hitting the target, and produce a defensible 30-60-90 day action plan ordered by probability lift, not gap size.

  Self-containedInstructions only1,662 words
• gtm-readiness-scan

  Run the GTM Readiness Scan as a 24-question Quick Scan or a 120-question Deep Scan interview, score 8 domains 0-5, compute the GTM Readiness Index, determine the company archetype, and produce a report with the weakest domains and a recommended focus.

  Reaches the webInstructions only2,194 words
• icp-agent

  Define, refine, and validate an Ideal Customer Profile using the Science of Scaling framework. Two modes: Quick Start (structured analysis from inputs) or Deep Dive (CRM data analysis with scoring model).

  Self-containedInstructions only806 words
• icp-objection-mapping

  Role-play skeptical ICP before writing campaigns to preempt objections

  Self-containedInstructions only1,881 words
• icp-scoring-dynamic

  Calculate dynamic ICP scores from enriched data using Claygent (Patrick Spychalski)

  Reaches the webInstructions only991 words
• ideal-customer-profiles

  Identify who a company serves based on their description and content

  Self-containedInstructions only296 words
• lead-scoring

  Design lead scoring models using firmographic, behavioral, and custom data signals

  Reaches the webInstructions only775 words
• linkedin-posts-summary

  Summarize a prospect's recent LinkedIn posts for personalized outreach

  Self-containedInstructions only506 words
• list-is-the-message

  Build segments where the "why" is so clear that messaging writes itself (Jordan Crawford)

  Reaches the webInstructions only715 words
• meeting-prep

  Pre-call research agent that produces a structured brief for any sales meeting. Combines web research with SPICED discovery prep and 6 non-negotiable discovery questions, customized per contact and company.

  Self-containedInstructions only748 words
• metaprompter

  Use AI to improve AI prompts by asking what context is missing (Eric Nowoslawski)

  Self-containedInstructions only866 words
• multi-product-detection

  AI prompt to detect multi-product company structure from public sources (Petra Hajal)

  Self-containedInstructions only707 words
• outcome-seller-scoring

  Turn call recordings into a per-rep SPICED coverage heatmap and a 50,000-run Monte Carlo of rep-level revenue. Scores every call on the 5 SPICED axes plus the 4 Outcome Seller moves, marks each SPICED axis Qualified / Partial / Gap against WbD's quote-or-Gap deal-review bar, reports coverage and qualified rate per dimension (org view and per-rep view), then simulates P10/P50/P90 revenue per rep per month, splits the team on the 80% ladder, ranks which lever closes the gap, and prices the coaching case by simulating the middle cohort at top-cohort coverage. Use for "score these calls", "rep scorecard", "SPICED coverage", "call archive diagnostic", "how much revenue is coaching worth", "Monte Carlo the reps", or a free call-assessment offer. Built on Winning by Design's SPICED and Jacco van der Kooij's Outcome Seller webinar (Aug 2026). Not for extracting competitive or handoff notes from calls (conversational-intelligence) or a quick SPICED score on a few transcripts (spiced-call-scorecard).

  Self-containedActs undeclared3,253 words
• pain-qualified-segment

  Build segments based on tension heuristics that indicate active pain points, not just firmographics

  Reaches the webInstructions only786 words
• permissionless-value-proposition

  Create independently valuable outreach by combining public data for actionable insights (Jordan Crawford)

  Reaches the webInstructions only827 words
• playbook-generator

  Cannonball GTM Playbook Generator. 6-phase intelligence-driven playbook workflow using EDP methodology and pain-based segmentation. ALWAYS execute all 6 phases sequentially (research, EDP analysis, segment scoring, data source discovery, play generation, scoring and assembly). NEVER generate a playbook in one shot.

  Self-containedActs undeclared2,896 words
• playbook-pitch-generation

  Generate company-specific pitch ideas and playbooks per account using AI (Patrick Spychalski)

  Reaches the webInstructions only907 words
• plg-company-detection

  AI prompt to detect PLG (product-led growth) companies from public sources (Petra Hajal)

  Self-containedInstructions only728 words
• pqs-pvp-messaging

  Write a PQS (Pain-Qualified Segment) or PVP (Permissionless Value Proposition) cold email from one signal or data point about a prospect. Covers only the message itself: the email that mirrors their situation (PQS) or hands them intelligence they could not build themselves (PVP). PQS and PVP are Jordan Crawford's concepts (Blueprint GTM); the voice rules follow Josh Braun. Not for campaign design, sequences, sending setup or segmentation, not for building the segment (pain-qualified-segment, list-is-the-message) and not for grading a finished draft (qa-checklist).

  Reaches the webInstructions only3,111 words
• pricing-strategy

  Infer a company's pricing model from their website or public information

  Self-containedInstructions only346 words
• product-complexity-detection

  AI prompt to detect product complexity from public sources (Petra Hajal)

  Self-containedInstructions only653 words
• prompt-engineering-rules

  Eric Nowoslawski's 5 rules for efficient AI prompts in Clay workflows

  Self-containedInstructions only906 words
• qa-checklist

  Ship-ready email standards and QA checklist for cold emails

  Self-containedInstructions only2,060 words
• recent-news

  Find and summarize recent news about a company for timely outreach

  Self-containedInstructions only397 words
• research-playbook

  10-minute research method with tools to find custom signals for cold emails

  Self-containedInstructions only2,312 words
• role-focus

  Analyze what a job title emphasizes to personalize outreach by role

  Self-containedInstructions only392 words
• saas-identification

  Determine if a company is a SaaS business based on their description

  Self-containedInstructions only440 words
• spiced-call-scorecard

  Score sales call transcripts (plain text, VTT, or SRT) against the five SPICED dimensions with verbatim-evidence discipline, then generate the follow-up questions that close the gaps before the next call.

  Self-containedInstructions only1,358 words
• tam-data-cost-estimate

  Triggers when a user wants to know what it will cost in data to map or source a market before building it, with phrasings like "what will this TAM cost", "price the list build", "data cost for 3,000 accounts", "estimate enrichment spend for this market", or "put a data budget in the proposal". Owns the row math (accounts, qualifying titles per account, priced fields, refresh) and the three numbers for a proposal, one-off build data, monthly run data and tool subscriptions, each as a range. Not for enriching records you already own in a CRM (crm-enrichment-cost-estimate), not for sizing the market itself, and not for sending volume or mailbox cost.

  Self-containedActs undeclared763 words
• win-loss-rewind

  Build an outcome-backward ICP. Works backward from real customer outcomes (won, lost, healthy, churned, expanded) to find the operational situations 6-18 months pre-purchase that actually predict who buys and stays, then ships a per-archetype scoring rubric validated on a 20% holdout. Use when you have an outcome-labeled customer list and need a situational ICP, not a firmographic filter. Based on Jordan Crawford's win-loss-rewind method (Blueprint GTM). Not for firmographic top-25%-by-revenue scoring (that is the anti-pattern this skill exists to stop), and not for a quick interview-based ICP (icp-agent).

  Reaches the webInstructions only2,205 words