# Lookatitude Guild Marketing Positioning > marketing-positioning - Skill: `tomevault-io/lookatitude-guild-marketing-positioning` (Agent Skill, multi-file: 2 files) - Install (CLI): `npx skillmds@latest add tomevault-io/lookatitude-guild-marketing-positioning` - Raw SKILL.md: https://api.skillmd.com/api/skills/tomevault-io/lookatitude-guild-marketing-positioning/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: Marketing & Growth - Author: tomevault-io (https://skillmd.com/u/tomevault-io) - Updated: 2026-09-17 - Page: https://skillmd.com/skills/tomevault-io/lookatitude-guild-marketing-positioning --- # marketing-positioning Implements `guild-plan.md §6.3` (marketing · positioning) under `§6.4` commercial principles: hypothesis-first (which ICP + category bet are we making?), success = measurable outcome (conversion, segment win rate), evidence = customer quotes, win/loss data, category analysis — not adjectives. ## What you do Produce a positioning statement tight enough that every downstream asset — site copy, launch, sales deck, campaigns — can be traced back to one source of truth. Positioning is a bet, not a description. Name the bet. - Define the ICP with firmographic + behavioral precision: role, company stage, trigger event, status quo pain. "SMBs" is not an ICP. - Name the market category explicitly — the frame the buyer uses to compare. Category choice determines competitive set. - List competitive alternatives, including status-quo ("doing nothing in a spreadsheet"), not just named vendors. - State the unique value as a capability + outcome pair, not a feature list. "Auto-reconciles in under a minute" beats "AI-powered reconciliation." - Provide proof: customer quotes, benchmark numbers, third-party data. Un-cited claims are vapor. - Write a "not-for" list — who this is explicitly wrong for. Vague ICPs dilute every message downstream. ## Output shape A markdown file at `.guild/runs//positioning/.md` with sections: 1. **For whom** — ICP with firmographic + trigger + status quo. 2. **The problem** — the pain the ICP feels today, cited. 3. **Category** — the frame / market we compete in and why. 4. **Unique value** — capability → outcome, 2–4 items, each provable. 5. **Competitive alternatives** — named vendors + status quo, with our frame vs each. 6. **Proof** — quotes, data points, case studies with source. 7. **Not for** — who this is explicitly wrong for. 8. **Open bets** — positioning hypotheses still unvalidated. Keep under ~300 lines. If it sprawls, the positioning is still fuzzy. ## Anti-patterns - Feature-led positioning: "We have SSO and webhooks" is not a position. - Vague ICP: "growing teams" is untargetable. - Category avoidance: "we're a new kind of thing" leaves the buyer with no frame. - Superlative soup: "best-in-class, enterprise-grade, AI-native" — zero information. - No "not-for": signals the team wants to sell to everyone, which means no one. - Un-cited proof: "customers love it" without a quote or number. - Confusing positioning with a tagline — the tagline is downstream output. ## Handoff Return the positioning doc path to the invoking `marketing` specialist. Downstream this seeds `marketing-launch-plan`, `marketing-campaign-brief`, `copywriter-voice-guide`, `sales-cold-outreach`, and `sales-discovery-framework`. If the positioning exposes a product gap, flag back to the `marketing` specialist for routing — this skill does not dispatch. --- > Source: [lookatitude/guild](https://github.com/lookatitude/guild) — distributed by [TomeVault](https://tomevault.io).