Grow Website

Grow a Website

Take a site that already has traffic and under-monetizes it, and turn it into a research-driven growth engine: more leads captured, more buyers persuaded, a sharper offer, and visitors who bring more visitors. This journey runs eight phases — the agent asks before it decides at every fork, and all state lives in docs/ so you can stop and resume across sessions. It sequences the work; it does not redesign the site on a hunch.

Core Principle

Most traffic problems are conversion problems in costume: capture and convert the visitors you have before buying more. This skill sequences the phases, asks the decision questions, and records every choice in docs/. The constituent skills carry the method — invoke them rather than improvising their frameworks. Research comes before optimization, and the offer comes before the page polish, always.

Journey Map

Operating Rules

1. Resume first. Before anything else, read docs/GROW-WEBSITE-PLAN.md and every artifact in the Journey Map. If the tracker exists, summarize the journey state in 3-5 lines and ask which phase to enter. Done when the user has confirmed an entry point. A journey with a tracker is resumed, never restarted.
2. Intake on first run only. No tracker: run the Intake below, then create docs/GROW-WEBSITE-PLAN.md with every phase statused pending | in-progress | awaiting-evidence | done | deferred: reason | skipped: reason. Done when the tracker exists and the user has confirmed the phase plan.
3. Phase entry. Announce: what the phase does, the decision it forces, the artifact it produces, rough effort. Offer proceed / skip / defer — phases marked GATE may be deferred, never skipped. Mark the phase in-progress on proceed. Done when the user chose.
4. Skill invocation and fallback. Load the phase's skill and use it: each phase's Invoke line names the skill by slug — use that skill to run the phase. If it is not available, offer: npx skills add wondelai/skills/<slug> --global. If the user declines, run the phase from its Brief — the minimum viable method. State which mode you are in.
5. In-phase decisions. Ask every question under "Decide with the user" — with concrete options and your recommendation. Record the choice in the tracker's Key Decisions. A decision made silently is a defect.
6. Phase exit. Present the draft artifact content for sign-off before writing. On approval: write or extend the docs/ files, update the tracker (status, Key Decisions, Next Actions). Done when the files are written and the phase row shows done.
7. Artifact discipline. Read before writing; create a file only if missing, otherwise extend — add or update your sections, preserve everyone else's. Files are UPPERCASE in docs/. Every recommendation lands as a checkbox or a table row with owner and priority. See references/artifact-templates.md when creating a docs/ file for the first time — create it from the full skeleton (all section headings), then fill the sections your phase names.
8. Every scarcity claim, testimonial, and share incentive must be true. When a persuasion tactic needs inventing evidence, replace it with a tactic that uses evidence you have. A countdown that resets, a "only 2 left" that never changes, or a fabricated review poisons every honest claim that follows it.

Intake

Ask these before creating the tracker:

1. What is the site, and what is its primary conversion action — signup, purchase, booking, lead form? (Anchors every phase and defines where the funnel ends.)
2. What traffic and conversion numbers do you have — sessions, bounce rate, conversion rate, worst-performing high-traffic page? (Gates Phase 1's funnel map and whether A/B testing can reach significance.)
3. What visitor research already exists — exit surveys, support tickets, chat logs, reviews? (Raw material for the O/CO table; if none, Phase 1 starts by gathering it.)
4. Do you already capture leads (list, lead magnet), and how well does it convert? (Gates Phase 2.)
5. What are you selling, at what price, and do visitors call it "too expensive"? (Gates Phase 6's offer rebuild.)
6. Is there existing positioning or messaging — a POSITIONING.md, a brand script, a one-liner? (Gates Phases 3-4; reuse rather than redo.)
7. What is your monthly traffic volume? (Decides Phase 1's testing approach: enough for bold A/B tests to reach significance, or lean on qualitative research and reason from principles.)

Skip heuristics: skip Phase 2 when a lead-capture funnel already converts well; skip Phases 3-4 when a stranger already passes the five-second test on the message; skip Phase 6 when the offer converts and price is not an objection; skip Phase 7 when there is no naturally shareable asset and referral is not a realistic channel.

Then create docs/GROW-WEBSITE-PLAN.md from the template and confirm the plan. Done when docs/GROW-WEBSITE-PLAN.md exists with every phase statused and the user has confirmed the plan.

Phases

Phase 1 — Discover why visitors do not convert (cro-methodology)

Purpose: Replace guesses about why visitors leave with evidence, and turn it into a ranked test queue.

Brief (fallback): Don't guess, discover. Map the funnel to find blocked arteries (high-traffic pages that drop off) and missing links. Research visitors on three axes: who they are, what blocks them (UX), what stops them (objections). Build an Objection / Counter-Objection table in the visitor's own words, place each counter at its point of friction, and inventory proof you have but do not show. Queue bold experiments — not button colors — scored by ICE.

Invoke: Use the cro-methodology skill with the primary conversion action, traffic numbers, and any research (exit surveys, support tickets, chat logs, reviews) from intake. Ask for a funnel map, an O/CO table sourced from real visitor language, an audit of the worst high-traffic page scored 0-10, and 3-5 ICE-scored experiment ideas with a hypothesis for the top two.

Decide with the user: (1) Which page is the blocked artery to attack first — highest traffic times worst conversion? (2) Is there enough research to source objections, or do we run exit surveys first (status awaiting-evidence)? (3) Which two experiments enter the queue — recommend the highest ICE that could plausibly double conversion, not a meek tweak.

Artifact: Extend docs/WEBSITE.md ## Audit Findings and ## Conversion Elements (the O/CO table with placement); docs/METRICS.md ## Funnel (stages, drop-off, benchmark, bottleneck); docs/EXPERIMENTS.md ## Experiment Backlog (ICE) and ## Experiment Cards for the top two. Update the tracker.

Done when: the funnel is mapped with the blocked artery named, WEBSITE.md holds an O/CO table with counters placed at their friction point, and two ICE-scored experiment cards exist with pre-committed primary, secondary, and guardrail metrics.

Phase 2 — Capture the 97% who are not ready to buy (scorecard-marketing)

Purpose: Keep the visitors who will not buy today with an interactive assessment instead of losing them at exit.

Brief (fallback): About 3% of a market is ready now; capture the rest. A scorecard or quiz converts 30-50% versus 3-10% for a PDF. Four steps: a landing page built on the 3 Cs (Clarity, Credibility, Connection); a questionnaire that captures the email before the questions, then 8-15 scored questions in 2-7 categories; a results page with dynamic content per tier that creates tension between where the visitor is and could be; a follow-up engine segmented by score.

Invoke: Use the scorecard-marketing skill with the audience and the primary conversion action. Ask for the concept hook (pressure-test several), the scored questions and their categories, the tiered results copy, the pre-question capture form, and tier-segmented follow-up emails.

Decide with the user: (1) The concept hook — pick a "moving toward" framing that taps the strongest dormant desire over a fear-based one. (2) How many tiers and where the score cutoffs sit. (3) Does an adequate capture funnel already exist, in which case skip?

Artifact: Extend docs/WEBSITE.md ## Lead Capture (scorecard funnel design: hook, questions, tiers, capture-first form); docs/MARKETING.md ## Nurture Sequences (the tier-segmented follow-up tracks). Update the tracker.

Done when: WEBSITE.md documents a scorecard that captures email before question one with dynamic results by tier, and MARKETING.md holds a separate follow-up track for each tier.

Phase 3 — Make the message clear before clever (storybrand-messaging)

Purpose: Ensure a stranger can tell what you do, for whom, and why to care within five seconds.

Brief (fallback): Make the customer the hero and the brand the guide — Yoda, not Luke. Run SB7: a Character who wants one thing; a Problem named at three levels (external, internal, philosophical — sell to the internal); a Guide with empathy and authority; a 3-4 step Plan; a direct and repeated Call to Action; the stakes of Failure; a picture of Success. Produce a one-liner: "We help [character] who struggle with [problem] to [solution] so they can [result]." Clear beats clever.

Invoke: Use the storybrand-messaging skill with the objections and proof inventory from Phase 1 — they are the raw material for the Problem and the Guide's authority — plus any existing positioning. Ask for a full brand script, a homepage hero rewrite, and 2-3 one-liner options.

Decide with the user: (1) Which internal problem to lead with. (2) Which one-liner a stranger could repeat after hearing it once. (3) The direct plus transitional CTA wording.

Artifact: Extend docs/POSITIONING.md ## Brand Script (StoryBrand), ## One-Liner, and ## Key Messages (surface | message | status). Update the tracker.

Done when: POSITIONING.md holds a completed brand script, a chosen one-liner, and key messages mapped to surfaces — and the hero passes the five-second test.

Phase 4 — Make the message stick in memory (made-to-stick)

Purpose: Make the clear message memorable so a visitor who is not ready yet still recalls you when they are.

Brief (fallback): Apply SUCCESs: Simple, Unexpected, Concrete, Credible, Emotional, Stories. Beat the Curse of Knowledge — replace jargon and abstraction with concrete, human-scale specifics ("orders arrive in 30 minutes, still hot"; "saves 16 hours a month"). Lead with the counterintuitive thing, not the expected one. Wrap the transformation in one specific customer's story: if I look at the one, I will act.

Invoke: Use the made-to-stick skill with the Key Messages and page copy from Phase 3. Ask for a Curse-of-Knowledge audit flagging every abstract claim, concrete rewrites with real numbers, and one flagship case study turned into a sticky story.

Decide with the user: (1) Which claims to make concrete first — the highest-traffic pages. (2) Which case study becomes the flagship story.

Artifact: Extend docs/POSITIONING.md ## Key Messages with a ### Sticky Rewrites subsection (concrete, de-jargoned versions), and docs/WEBSITE.md ## Page Briefs copy blocks. Update the tracker.

Done when: every abstract claim on the priority pages has a concrete rewrite recorded, and one sticky customer story exists.