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/output-format.md
View on GitHub# Output format
List mode only. Load at list-mode steps 5–6 (rows) and when writing the list report.
## `list.csv` columns (in this order; one row per person)
`row_type, list_source, rank, tier, value_tier, value_evidence, fit_score, fit_reason, segment, persona_role, first_name, last_name, title, title_source_url, title_seen_date, profile_url, email, email_candidate, email_status, email_source, other_contact_route, company, domain, account_id, employees, employees_source, industry, person_location, region_flag, signal_1, signal_1_url, signal_1_date, signal_2, signal_2_url, signal_2_date, hook, hook_url, hook_date, warm_path, notes, sources, retrieved_date`
- `email` is filled **only** after a checker passed the address; `email_candidate` holds published or pattern-built addresses until then (contacts-and-compliance.md).
- `list_source`: where the row came from — `database`, `switch-intent`, `hiring`, `new-leader`, `rival-customer`, `event`, `developer-activity`, `ads`, `champion`, `directory`, `their-list` (score-a-list), `watch`.
- `region_flag`: `UK-ltd` (B2B email with opt-out), `EU-UK:<country>` (check national rule), `UK-soletrader: consent needed`, `CASL`, `AU-NZ`, `US`, `other:<country>` (check local rule). `title_source_url` may be a `search-headline (<date>)`.
- `row_type`: `person`, `account`, or `account-to-find` (passed the gates, no named person yet — "people to find"; write-outreach skips it) (chains' portal or buying-office row, with any confirmed buyers as person rows under the same `account_id`; independent shops with a published inbox, IG, or — UK sole traders — a visit/letter address).
- Retail lists add `doors, price_tier, stocks_competitor, review_window` after `industry`.
- Tier C rows go to `dropped.csv` with their reason; after a verifier pass, catch-alls move to `catch-all.csv`.
- `warm_path`: only from what the human supplied (connections export, investor list, customer champion); blank otherwise.
- Plain CSV, UTF-8, header row, no merged cells — imports into common CRMs and outreach tools.
- Beside `list.csv` in `<D>-<list>/`: `segments.csv` (`segment, name, pain, pain_source_url, proof, ask, default_opener, rows` — the angles write-outreach reads) and, after a checker pass, `catch-all.csv` (the same columns as `list.csv`).
- `dropped.csv`: `company, domain, person (if any), stage dropped, reason`.
## The list in the report
List mode delivers its own short list report (list-mode.md § In the report; rules: the `write-report` skill (`../../write-report/SKILL.md`)); when an audience report was also run, the list is its last section. Write it for the founder: plain words, no codes. Say segments and tiers in words ("agencies hiring sourcers now", "top accounts"), never "S1" or "V1" — those labels live only in the CSV.
```
## The list: 41 people to pitch at 23 agencies
**Send first:** the six largest agencies hiring sourcers right now — each posted the role in the last 14 days ([links](…)).
- Found 41 of the 50 asked: the market is thinner than it looks — about 140 agencies match in the US and UK.
- Email: 18 printed on the agency's own site, 15 built from a pattern we saw there, 8 reachable only on LinkedIn or by phone. None is checked yet — run them through your email checker first.
- Angle for the hiring group: *their pain* — every new hire adds manual search; *proof* — you'd need a customer's hours saved first; *ask* — "worth 15 minutes before the new sourcers start?"
- Before you send: EU/UK rows need the local rule checked; Canada and Australia only to addresses printed on their own pages; UK sole traders by letter or visit, never cold email.
- The full list with sources: [list.csv](…)
```
## Sample rows (FICTIONAL — the bar, not data; never reuse these values)
| tier | score | person | company | fit_reason | signal | hook | email_status |
|---|---|---|---|---|---|---|---|
| A | 88 | Dana Ruiz, Controller | Brightfold (brightfold.example), 640 staff, US | Gates passed: US, 640 staff (company page 2026-10-14), not a customer. Account: wholesale distributor mid-band (+20). Persona: Controller owns AP, team page (+20). Tech: NetSuite named in a finance job post, not AP's own (+8). Timing: hiring 2 AP specialists, posted 2026-10-02, live 2026-10-15 (+40). | "2× AP Specialist roles open since 2026-10-02" [url] | Posted 2026-09-28 about month-end close taking 9 days [url] | published (brightfold.example/team) |
| A | 81 | Sam Okafor, VP Finance | Harlow Freight, 900 staff, US | Gates passed… Account: core band (+22). Persona: economic buyer (+20). Tech: NetSuite on the integrations page (+12). Timing: started as VP Finance 5 weeks ago, from a NetSuite shop (+27). | "New VP Finance since 2026-09-11" [url] | Spoke on a CFO podcast about "killing paper invoices" 2026-09-19 [url] | pattern-guess (first.last, evidenced by jo.lin@harlowfreight.example on the press page) |
| B | 48 | Lee Tan, Finance Ops Manager | Corvel Goods, 310 staff, US | Gates passed… Account: lower band (+18). Persona: champion (+20). Tech: NetSuite in a job post (+10). Timing: none live (0) — tier B on fit ≥45/60. | none | none found | none — LinkedIn route only |
What makes these good: every number is sourced and dated; the reason names the gates and the points; the signal says why *now*; the hook is a fact tied to the pain; the email status says exactly how much to trust the address.
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 |