gooseworks-ai/goose-skills / skills/marketing/where-to-start
where-to-start
VerifiedFirst-move advice for a broad growth or marketing ask with no job named: looks at what GooseWorks already knows and at the site, asks a few short questions in one message, names the one thing in the way, and sends the marketer to the right next job. Use when the marketer says "help me grow my brand", "I need to grow my business", "how do I get more customers / sales", "help me with marketing", "where do I start", "sales are flat", "what should I focus on this month", or "I'm back, what's next?". Recommends only; never runs the work. Not for a named job like "make me ads" or "grow my Instagram" (go straight to it); not for a campaign with a launch, budget or date — use plan-campaign; not for why this week's numbers moved — use analyze-performance.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 12.4 KB |
| references/base-rates.md | markdown | 5.1 KB |
| references/output-shape.md | markdown | 7.0 KB |
| references/playbook-map.md | markdown | 19.8 KB |
| references/questions.md | markdown | 5.1 KB |
| references/situation-lens.md | markdown | 9.7 KB |
| skill.meta.json | config | 3.2 KB |
| tests/test-by-hand.md | markdown | 9.9 KB |
references/questions.md
View on GitHub# The five situation questions
Load at steps 1 and 2. Five questions cover what a senior advisor needs before naming a first move. Read the brain first; ask only what it cannot answer. Each question exists because its answer can change the recommendation.
| # | Question (plain words) | Why it changes the answer | Often already in the brain |
|---|---|---|---|
| 1 | **What do you sell, to whom, and where do people buy it?** (a store / an app / software for businesses / a local business / a service / something else, plus a link) | Company type, sales motion and where it is sold decide which playbooks are even relevant. | `brand_read` summary (description, `brandType`), products, kit `audience`; the short look at the homepage and store page |
| 2 | **Where are you right now?** (a) not launched yet; (b) launched, few or no paying customers; (c) steady sales, want more; (d) it was growing, now it's flat or falling. **And who are your customers today, and how did they find you?** (The proxy for fit — (b) are first users active every week, and would they pay? (c) what share buy again? (d) what changed when it flattened? — goes in the follow-up round, unless an ask is spare.) | Stage, the channel that already works, and the fact that separates a fit problem from a demand or conversion problem. | `ad_library_ad_count` in the brand list (own ads live), the rating line among research claims, organic posts, campaigns (`campaign_read`) |
| 3 | **What have you tried, how did it go, and what were you thinking of trying next?** Plus anything you already have: an email list of about how many, a following, a network, press, partners — **and notes from user or cancel calls: paste them.** (Ask this before ever recommending "talk to N users": the notes may already answer it.) | Avoids repeating failures; surfaces unfair advantages that beat any cold channel; the idea they bring ("maybe run some ads?") must be answered, often under "not yet". | the one `knowledge_search` across evidence, learnings, documents and creatives (own ads, offers, results), accounts, learnings |
| 4 | **What can you put in each month?** Money: $0 / under $1k / $1–5k / $5k+. Your time: about 2 / 5 / 10+ hours a week. Who else does the work, if anyone. | Feasibility. Money without time points one way, time without money another. | onboarding `ad_spend` band (a band, not a number) |
| 5 | **What would make the next 90 days a win, and is a big date coming** (a launch, a season, a fundraise, an event)? **Where do you think it's stuck?** | The goal and deadline set the urgency and the order; the founder's own guess surfaces the constraint (and is often a symptom). | `brand/situation.md` from an earlier visit, onboarding `goals` |
## How to ask
- **Open with what you know.** "Here's what I already know about you: … Correct anything." Then ask only the gaps.
- **One message, numbered, with options** plus "or say it your way". **At most five asks in total** — count every separate thing you ask, not the numbered lines: five numbered questions holding fourteen asks is fourteen. Each numbered question is one ask the founder can answer with one pick or one line; anything more waits for the follow-up round. Pick the five whose answers could change the recommendation; drop the rest. Questions 2, 4 and 5 belong to the founder and are always asked on a first visit, however much the brain holds; a rich brain only lets you skip question 1.
- **State, don't ask, what you can work out:** the season and the category's big moments from today's date (e.g. a DTC brand in October: "Black Friday is about six weeks out"), the stage when the brain already shows it.
- **Plain words.** No CAC, ICP, ROAS, MER, PMF in the questions.
- **"I don't know" is an answer.** Offer a proxy ("roughly how many orders last week?", "who were your last three customers?"); accept ranges. If the unknown decides the recommendation, make the first move the one that answers it, or give a conditional answer ("if most of your customers came back for a second order, start here; if not, start there"). State any assumption in one line.
- **At most one follow-up round,** and only for a fact that would flip the recommendation. One to ask when it does: the founder calls a group "not our audience" (e.g. people who left over price) while the best customer they describe is the most price-sensitive (solo founders) — ask who the churners actually were before anything is saved.
- **Echo the situation back** in one line before recommending, so a misread is caught before the advice.
## Never
- Ask what the founder's own message already answered, or question 1 when the brain and site answer it.
- Treat a plan, brief or research saved on the brand as the founder's answers. It came from another session; name it, with its date, and ask whether it still stands.
- Turn it into a form or more than five questions.
- Ask for passwords, account access or documents. The recommended playbook asks for what it needs.
- Research beyond the short look (at most three of their own public pages): no competitors, ad libraries or paid data. The founder's words, the brain and those pages are the only inputs.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |