gooseworks-ai/goose-skills / skills/marketing/research-audience
research-audience
VerifiedAudience research: who the buyers are, in their own words, where they gather and how to reach them — plus named prospect lists when asked. Use when the marketer says "research our audience", "who should we target", "mine our reviews", "why do people buy us or leave", "where does our audience hang out"; or for lists: "find 50 right-fit leads with emails", "find retailers or accounts to pitch", "score these leads / who do we call first", "who's looking for an alternative to X", "tell me when a past champion changes jobs". Reads the brand's own customer data first and delivers a readable report saved to the brand. Never contacts anyone. Not for market size or trends — use research-market; not for writing the messages — use write-outreach.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 18.7 KB |
| references/audience-research.md | markdown | 15.9 KB |
| references/contacts-and-compliance.md | markdown | 9.7 KB |
| references/data-sources.md | markdown | 12.3 KB |
| references/fit-and-scoring.md | markdown | 7.6 KB |
| references/list-by-company-type.md | markdown | 4.5 KB |
| references/list-mode.md | markdown | 7.5 KB |
| references/list-sources.md | markdown | 9.9 KB |
| references/output-format.md | markdown | 5.3 KB |
| references/quote-bank.md | markdown | 6.3 KB |
| references/run-memory.md | markdown | 5.2 KB |
| references/score-a-list.md | markdown | 7.6 KB |
| references/signals-and-hooks.md | markdown | 6.1 KB |
| references/watches.md | markdown | 4.7 KB |
| skill.meta.json | config | 6.3 KB |
| tests/test-by-hand.md | markdown | 8.9 KB |
references/signals-and-hooks.md
View on GitHub# Buying signals and personalisation hooks
Load at list-mode step 4 (signals) and 5b (hooks, after people). A signal is evidence of *timing*; fit says who, timing says why now.
## Evidence standard (non-negotiable)
Every signal and hook in a row is: **one-line fact · URL · publish date · retrieved date · why it matters now.** No URL, no signal. Re-check that job posts are still live on the day you hand the list over.
## Signals and freshness windows (strength anchors under Timing points)
Approximate windows from outbound practice; older than the window = drop it.
| # | Signal | Fresh for | How to see it with what we have |
|---|---|---|---|
| 1 | A past champion/user of the seller moved to a new in-ICP company | 0–120 days | Only if the human supplies champions (customer contacts list); check their public profile for the new employer |
| 2 | New leader in the buying role | first 30–90 days | Public profile shows the new role; announcement posts; press releases via search |
| 3 | Live job post for the role the product replaces/supports, or naming the pain/tool | open now, posted ≤45 days | Search job boards and the company careers page (e.g. queries like `"<company>" "AP specialist" site:boards.greenhouse.io OR site:jobs.lever.co`) |
| 4 | Public statement of the pain (post, talk, podcast, earnings call) | ≤60 days | Search the person's recent public posts; post search on LinkedIn/X; podcast/YouTube transcripts |
| 5 | Tech change: adopted a complementary tool, dropped a competitor | ≤90 days | Job posts naming tools, integration/partner pages, public code repos for dev tools |
| 6 | Expansion: new market, location, product line, store opening | ≤90 days | Press, the company's own news page, location pages |
| 7 | Running paid ads now (budget + active marketing) | live | Meta / Google / LinkedIn / TikTok ad libraries. Strong for selling *to marketers* (agencies, creative, martech) |
| 8 | Funding round | ≤90 days | Company page funding block (often stale — confirm with a dated press source). **Weak alone**: everyone emails newly funded companies; pair it with another signal |
| 9 | **Switch intent**: the person wrote "alternative to <rival>", "moving off <rival>", "<rival> too expensive" (list-sources.md, nine kinds) | ≤30 days (best ≤14) — one window everywhere | Web search, Hacker News search, public forum threads — the author only, identifiable at a fitting company |
| 13 | **Rival event**: a rival's price rise, acquisition, sunset, outage or big policy change — every one of its public customers becomes a switch candidate | ≤60 days from the event | News and the rival's own changelog or pricing page via web search; then its public customers (row 10) |
| 10 | Rival's public customer (case study, testimonial, logo wall) | case study ≤24 months; still using it | Rivals' customer pages; a recent job post naming the rival's tool confirms |
| 11 | Speaker, workshop host, sponsor or exhibitor at a relevant event | 90 days back, 180 ahead | Public event, speaker and sponsor pages — never attendee lists |
| 12 | Developer adoption: 3+ people at one company active on rivals' or category repos (issues, PRs, forks) | ≤90 days | Public GitHub API; an account signal — list the buyer, not the developers |
**Selling to retail and wholesale buyers** (consumer brands): category review / line-review calendar for the retailer (the retailer's published vendor calendar; otherwise a decision for the human), new store openings, the buyer's own posts about the category, trade-show exhibitor lists (never attendee lists), competitor brands newly stocked, a retailer's "local makers" or new-brand programme.
**Selling to local businesses / partners**: new opening, renovation, new owner, review-volume jump, running local ads, hiring.
**Noise — never count as a signal**: "growing fast", headcount growth with no function named, awards, follower counts, generic intent-topic surges, "uses Google Analytics", funding older than six months, work anniversaries, "visited our website" (we cannot see it).
## Timing points (for the 40-point block) — judgment, with anchors
Score timing by judgment, and write the reason in the row. Anchors so runs stay comparable:
- **Strength.** The strongest signals (a past champion who moved, switch intent in their own words, a new leader in the buying role, a live job post for a role we replace) can carry the full 40 when fresh; developer adoption, an event talk on the pain and a rival's recent case study sit with the middle signals; middle signals (a public statement of the pain, a tech change, an expansion) top out around 28; weak ones (running ads, funding) around 16. Retail: a category-review window open now is strong; a store opening, a competitor newly stocked or a new-brand programme is middle; a trade-show buyer list is weak. Local and partners: a new opening or new owner is middle.
- **Recency.** A signal early in its window counts fully; one near the end of its window counts for much less — a new leader in week 2 is a different conversation from one in week 11. Say which.
- **Two signals.** Independent signals add a little, never double; cap 40.
- Worked: a job post for the pain, 14 days old → near 40 ("live, posted 14 days ago, still open"); a new VP 11 weeks in → low teens ("late in the window — budget likely set").
When outcomes come back, re-tune these anchors from what actually replied (the outcomes ledger) and say what changed.
## Personalisation hooks — one per person
A hook is a *fact* the human's first line can build on, not copy. It must connect to the pain, not just prove research happened.
Worth it (with URL): something they published or said in the last ~60 days; their own job post; their recent move; a shared verifiable connection (former employer, same integration, mutual customer the human named); a specific company event.
Worthless: "saw you went to State U", compliments on the website, "hope you're well", invented insight, anything older than six months.
**Banned (creepy or unlawful)**: family, health, religion, politics, home location, personal (non-professional) social accounts, birthdays, inferred salary, photos. If a hook needs one of these, there is no hook — write "none found".
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |