ICP Website Audit
End-to-end website audit through ICP eyes. Builds synthetic personas (if they don't already exist), runs a structured scorecard review of the client's site, then runs a head-to-head competitive comparison against top competitors. Produces a single consolidated report with persona feedback, competitive positioning, and prioritized recommendations. The complete "how do our buyers actually experience our site vs the competition?" workflow.
by Athina AInpx gooseworks install --all # then, in Claude Code, Cursor, or Codex: /gooseworks use the icp-website-audit skill
ICP Website Audit
The complete "how do buyers experience our site vs the competition?" workflow. Chains persona building, website evaluation, and competitive comparison into a single end-to-end audit.
Quick Start
Run an ICP website audit for [company]. Site: [url]. Compare against [competitor 1] and [competitor 2].With existing personas:
Run an ICP website audit for [client]. Personas already exist. Compare against [competitor urls].Inputs
| Input | Required | Source |
|---|---|---|
| Company name | Yes | User provides |
| Company URL | Yes | User provides |
| Competitor URLs | Yes (1-3) | User provides, or discovered in Phase 1 |
| Client context file | Optional | clients/<client>/context.md |
| Existing personas | Optional | clients/<client>/personas/personas.json |
Step-by-Step Process
Phase 1: Persona Setup
Check if personas already exist:
clients/<client>/personas/personas.jsonIf personas exist: Load them, confirm they look current, and list them for the user. Skip to Phase 2.
If no personas exist: Run buyer-persona-generator:
- Research the company — what they sell, who they serve, pricing model, stage
- Identify 4-6 ICP segments from website, case studies, reviews, job postings
- Build detailed synthetic personas with full profiles
- Save to
clients/<client>/personas/
Output from this phase:
clients/<client>/personas/personas.json(machine-readable)clients/<client>/personas/personas.md(human-readable)clients/<client>/personas/segments.md(summary table)
Phase 2: Website Scorecard Review
Run icp-website-audit in scorecard mode against the client's own site.
- Crawl the client's site — homepage, pricing, product, solutions, about, case studies, blog, docs
- Check external presence — search results, review sites, social proof
- Run each persona through the site, scoring on:
- First Impression (1-10)
- Messaging Relevance (1-10)
- Trust & Credibility (1-10)
- Clarity & Navigation (1-10)
- Objection Handling (1-10)
- Overall (1-10)
- Cross-persona synthesis — consensus issues, segment-specific gaps, messaging disconnects
Output from this phase:
- Per-persona scored reviews
- Cross-persona analysis
- Initial recommendations
Phase 3: Competitive Head-to-Head
Run icp-website-audit in head-to-head mode against each competitor.
For each competitor (1-3):
- Crawl the competitor's site — same pages as the client site
- Run each persona through BOTH sites
- Per-persona comparison:
- Quick takes on each site
- Dimension-by-dimension scoring for both
- "If I had to choose today..." verdict
- What the client should steal from the competitor
- What the client does better
- Cross-persona competitive summary:
- Where the client wins across personas
- Where the competitor wins across personas
- At-risk segments (personas who lean toward competitor)
- Priority competitive moves
Output from this phase:
- Per-competitor, per-persona head-to-head reviews
- Competitive positioning map
- At-risk segments identified
Phase 4: Consolidated Report
Merge all findings into a single audit report. This is the deliverable.
Synthesis work:
- Combine scorecard + competitive findings — scorecard shows absolute quality, head-to-head shows relative positioning. Both perspectives matter.
- Identify the "so what" — what are the 3-5 moves that would make the biggest difference?
- Rank recommendations by:
- How many personas it affects (breadth)
- How much it would change conversion likelihood (depth)
- Whether competitors already do it better (urgency)
- Estimated effort (feasibility)
- Flag at-risk segments — ICPs where competitors are winning. These need immediate attention.
Output
Save to the current working directory as icp-website-audit.md (or user-specified path).
Also save the sub-reports to the current working directory:
personas/(persona assets, reusable)icp-reviews/<date>-scorecard.md(site review)icp-reviews/<date>-head-to-head-<competitor>.md(per competitor)
Consolidated Report Template
# ICP Website Audit: [Company Name]
**Date:** [Date]
**Website:** [url]
**Competitors:** [competitor 1], [competitor 2], [competitor 3]
**Personas used:** [N] (from `clients/<client>/personas/`)
---
## Executive Summary
[5-7 sentences. How do buyers actually experience this site? What's working, what's
broken, where are competitors winning? Top 3 things to fix and why.]
### Scorecard Overview
**Average score across personas: [X/10]**
| Persona | Segment | Own Site Score | Would Convert? |
|---------|---------|:-:|:-:|
| [Name] | [Segment] | [X/10] | [Yes/Maybe/No] |
| ... | ... | ... | ... |
### Competitive Overview
| Persona | Own Site | [Comp 1] | [Comp 2] | Pick |
|---------|:-------:|:--------:|:--------:|:----:|
| [Name] | [X/10] | [X/10] | [X/10] | [Winner] |
| ... | ... | ... | ... | ... |
**Personas at risk:** [Names of personas who preferred a competitor — these represent ICP segments you may be losing]
---
## Part 1: ICP Segments & Personas
### Segments Identified
| # | Segment | Description | Priority |
|---|---------|-------------|----------|
| 1 | [Name] | [One line] | Primary |
| 2 | [Name] | [One line] | Primary |
| 3 | [Name] | [One line] | Secondary |
| ... | | | |
### Persona Summaries
[2-3 sentence summary of each persona — who they are, what they care about, how they evaluate. Link to full profiles in `personas/personas.md`.]
---
## Part 2: Website Scorecard
### Score Matrix
| Dimension | [P1] | [P2] | [P3] | [P4] | [P5] | Avg |
|-----------|:----:|:----:|:----:|:----:|:----:|:---:|
| First Impression | | | | | | |
| Messaging Relevance | | | | | | |
| Trust & Credibility | | | | | | |
| Clarity & Navigation | | | | | | |
| Objection Handling | | | | | | |
| **Overall** | | | | | | |
### Persona-by-Persona Highlights
For each persona, include:
- **[Persona Name]** ([Score]/10) — [One sentence summary]. Liked: [key positive]. Frustrated by: [key negative]. Blocker: [#1 thing stopping conversion].
### Cross-Persona Findings
#### Consensus Issues (All Personas)
1. [Issue] — [Why it matters]
2. ...
#### Segment-Specific Gaps
| Gap | Personas Affected | Impact |
|-----|------------------|--------|
| [Gap] | [Names] | [H/M/L] |
#### Messaging Disconnects
| Site says | Buyers say | Personas |
|-----------|-----------|----------|
| "[quote]" | "[quote]" | [Names] |
---
## Part 3: Competitive Analysis
### vs [Competitor 1]
**Overall: [Company wins / Competitor wins / Split]**
| Dimension | [Company] | [Competitor] | Winner |
|-----------|:---------:|:------------:|:------:|
| First Impression | [avg] | [avg] | |
| Messaging | [avg] | [avg] | |
| Trust | [avg] | [avg] | |
| Clarity | [avg] | [avg] | |
| Objections | [avg] | [avg] | |
| **Overall** | **[avg]** | **[avg]** | |
**Where competitor wins:**
- [Specific advantage, which personas care]
**Where we win:**
- [Specific advantage, which personas care]
**Personas who'd choose competitor:** [Names — these are at-risk segments]
---
### vs [Competitor 2]
[Same structure]
---
### Competitive Position Map
| Persona | Best Experience | Worst Experience | Notes |
|---------|:-:|:-:|---|
| [P1] | [Company/Comp] | [Company/Comp] | [Why] |
| [P2] | ... | ... | ... |
### At-Risk Segments
[Personas who preferred competitors. For each: which competitor, why, and what it would take to win them back.]
### Competitive Advantages to Amplify
[Things the company does better than ALL competitors — lean into these.]
### Gaps to Close
[Things competitors do better across most personas — neutralize these.]
---
## Part 4: Prioritized Recommendations
### Tier 1: High Impact, Do Now
These affect the most personas and/or address competitive gaps.
1. **[Recommendation]**
- **Why:** [Evidence from scorecard + competitive analysis]
- **Personas affected:** [Names]
- **Competitive context:** [Do competitors already do this better?]
- **Expected impact:** [What changes — conversion, trust, clarity]
- **Effort:** [Low/Med/High]
2. **[Recommendation]**
...
### Tier 2: Medium Impact, Plan For
3. **[Recommendation]**
...
### Tier 3: Segment-Specific Optimizations
These improve experience for specific ICP segments.
5. **[Recommendation]** — For [persona/segment]
...
### Quick Wins
- [Small change] → Benefits [persona]
- [Small change] → Benefits [persona]
---
## Appendix
### A. Full Persona Profiles
→ `clients/<client>/personas/personas.md`
### B. Detailed Scorecard Review
→ `clients/<client>/icp-reviews/<date>-scorecard.md`
### C. Detailed Competitive Reviews
→ `clients/<client>/icp-reviews/<date>-head-to-head-<competitor>.md`
### D. Persona Data (JSON)
→ `clients/<client>/personas/personas.json`Parallelization
Some phases can run in parallel to save time:
Phase 1: Build personas (if needed)
↓
Phase 2 + Phase 3 can partially overlap:
- Crawl all sites (client + competitors) at once
- Run scorecard first (client site only)
- Then run head-to-heads (reuse client crawl data)
↓
Phase 4: Consolidate (needs all prior phases)If personas already exist, the whole workflow is just:
Crawl all sites → Scorecard → Head-to-heads → ConsolidateTips
- Start with 1-2 competitors, not 5. Each competitor adds a full head-to-head evaluation pass. Start focused, add more if needed.
- The at-risk segments are the most actionable finding. If a persona consistently prefers a competitor, that's an ICP segment you're actively losing deals in. That should drive urgency.
- Reuse this quarterly. Run personas once, then re-run the audit after site changes or quarterly to track improvement. Compare scores across dates.
- Combine with SEO content audit for the full picture. ICP audit tells you "does the messaging work for buyers?" SEO audit tells you "are buyers finding you in the first place?" Together they cover the full funnel.
- This makes a great client deliverable. The consolidated report is structured for sharing. Use
content-asset-creatorto make it into a polished HTML report. - Don't skip the competitive analysis. The scorecard alone is useful, but the head-to-head is where the real insights are. Buyers always compare — your audit should too.
Dependencies
- Skills:
buyer-persona-generator,icp-website-audit - Web search capability (for research, external presence checks)
- Web fetch capability (for reading website pages)
- No API keys or paid tools required
Cost
Free — no paid APIs. All research and evaluation is done via web search and web fetch.
What's included
clients/<client>/personas/personas.json (machine-readable)clients/<client>/personas/personas.md (human-readable)clients/<client>/personas/segments.md (summary table)Render Model Comparison Grid
Render a 'model comparison grid' video from a config — a fal-style "same prompt, N contenders" showcase — a dark real-DOM stage where per beat a monospace prompt fades in centered, docks to a small top strip, then a labeled 2-4 panel grid (static images OR muted video clips, mixable per cell) staggers in and holds for comparison, plus a minimal end card — frame-stepped via Playwright (video cells are frame-seeked deterministically) and encoded with FFmpeg. Deterministic assembly, FREE (cell media comes from create-image-fal / create-video-fal, music from create-music-elevenlabs), text stays pixel-crisp. Use for the model-comparison-grid format.
Render Offer Ad
Render a punchy ~12s vertical (9:16) music-only direct-response OFFER ad as a 4-beat kinetic-typography film — HEADLINE slam → real PRODUCT drop → CLAIM/proof → CTA pill — from one config of copy slots, a real product photo, a brand palette, fonts, bpm, and beat split. DETERMINISTIC + FREE (a bundled Remotion project; springs + interpolate, no AI-gen for visuals). Backgrounds are engine gradient divs off the palette, props are inline SVG, the ONLY composited bitmap is the REAL product photo (objectFit:contain, never stretched), and ALL headline/claim/CTA/URL/wordmark text is typeset in the engine — never AI-rendered (the format's credibility guard). A driver binds the config to Remotion input props, renders the 9:16 master, and derives a 1:1 center-crop with ffmpeg. Two gating checks run before render (claim verbs must match the product's physical format; the claim beat needs an edge-entry mechanism prop). Use for the motion-graphics-offer-ad format.
Render Myth Vs Fact
Assemble a myth-vs-fact kinetic-typography explainer video ad (≈29.5s, 9:16) from N myth/fact pairs + hook / turn / punch copy + palette + a brand end-card PNG + a VO track — a hook, 3 red-strike MYTH cards that flip to teal-check FACT cards (per-line strikethrough that crosses EVERY wrapped line), a "what actually works" turn, an optional proof reveal, a punch line, and a static end card. DETERMINISTIC assembly with ZERO AI-gen visuals — HTML hyperframes rendered frame-exact via Playwright (`window.renderAt(t)`, animation a pure function of beat-local time), Whisper beat-snap to VO word onsets, concat at a uniform fps, karaoke `.ass` captions burned last (suppressed on the proof + end-card beats), and a VO + optional music mix (music −20 dB, `amix normalize=0`, tail fade). FREE (Python + Playwright + ffmpeg); the recipe supplies the copy / palette / end-card / VO and gates the paid VO / music / Whisper calls to their own capabilities. Use for the myth-vs-fact format.