Skip to content

gooseworks-ai/goose-skills / skills/marketing/work-with-creators

work-with-creators

Verified

Real creators and influencers, researched, priced and briefed. Use when the marketer says "find influencers", "which creators are worth paying, at what price?", "find real UGC creators to film for our ads", "who should we seed product to", "is this creator's engagement real", "what should I pay this creator" or "write a creator brief". Free research first; one credit yes only to vet deeper. Drafts only: contacts nobody, agrees nothing, moves no money. Not for AI-made UGC-style video — use goose-video; if unsure, ask: a real person filming, or AI-made?

references/wave-criteria.md

View on GitHub
# Wave-1 success criteria

Load at step 6. A creator programme without a yardstick never learns: every creator "did fine". Before anyone is contacted, write down per role what a win is, when it is judged, and what happens next. The human reports results; the outcomes ledger compares them with these criteria and the next run reweights the roster (roles, tiers, price bands, vetting weights).

**Time the wave to the buyer's calendar, not to our launch.** Posts land when buyers are deciding: B2B tool buyers plan in January (Q1 budgets) and are heads-down through Black Friday to year-end, so a 1–10 December wave for ad-tool buyers is the worst window — move it to January. DTC gifting lands before the buying season (weeks before a holiday, not on it). Write the reason next to the dates in `wave-1.md`.

## Per role (approximate starting points — set them from the brand's own numbers when it has them)

| Role | Win metric | Read window | Renew / cut |
|---|---|---|---|
| **Reach (paid or gifted)** | Post views ≥ the creator's organic median; saves + shares in their usual range; comments with purchase intent ("where do I get this?") | 7 days after posting | Renew if views ≥ median **and** intent comments appear; cut if under half the median or off-brief |
| **Seeding (gift-only)** | Post rate across the wave (expect ~20–40%); positive mentions | 21 days after delivery | Promote the posters with above-median views to paid; drop non-posters from the next wave |
| **Affiliate / hybrid** | Tracked orders on `[CODE]` / `[LINK]` vs the fee (first-order gross profit ≥ fee, or the walk-away logic in pricing §6) | 30 days | Renew if the fee paid back; renegotiate to more commission if close; cut if not |
| **UGC-for-ads** | When run as an ad: hook rate / thumb-stop and CPA vs the brand's target | 14 days of spend | Re-book creators whose asset beat the account's median CPA; cut those in the bottom third. **No ad history (new brand):** compare within the wave — hook rate and CPA against the wave's median, labelled "starting point" |
| **B2B voice / newsletter / podcast** | Clicks or demo requests on the tracked `[LINK]`; qualified replies; cost per demo vs target | 14 days (posts) · 30 days (newsletters, podcasts) | Renew if cost per demo ≤ target; cut if no qualified response. **No target yet:** a starting point the human confirms — e.g. at least 2 qualified demo requests or 5 qualified replies per placement — labelled "starting point" |

## What goes in `creators/<date>/wave-1.md`

- One block per role in the wave: the win metric with a number (from the brand's figures, or labelled "starting point"), the read window with dates, the renew / cut rule.
- The placeholders the human must create before posting: `[CODE]` and `[LINK]` per creator (a code and a tracked link each — the playbook never issues them).
- Who reports what, when: "After <date>, paste views / orders / replies per creator and I'll score the wave."
- The roster's `wave` column = `1`; `wave_result` is filled from the human's report (`win`, `partial`, `miss`, `no-data`).

## Read-back (next run, via the outcomes ledger)

Compare each pick's result with its criterion; say what the numbers support (which role and tier worked, whether price bands were right, which vetting signal predicted a win), with "directional only" under ~5 creators per group. Propose the changes; the human's yes before the plan or bands change.
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: 115kUpdated Claude CodeCodexVerifiedInstall
add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation.Stars: 115kUpdated Claude CodeCodexVerifiedInstall
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: 115kUpdated Claude CodeCodexVerifiedInstall
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: 115kUpdated Claude CodeCodexVerifiedInstall