Conversion Rate Optimization skills
Conversion Rate Optimization Skills-md for Claude & Codex
Conversion rate optimisation skills are SKILL.md files that turn Claude, Codex and other AI agents into careful CRO specialists. They audit the whole journey, run conversion research, improve landing pages, forms, pricing pages, checkout and sign-up flows, handle popups, social proof, copy and mobile, plan tests that fit your traffic and report results leadership can trust. Every finding becomes a hypothesis with evidence and a metric. All 13 are below: copy one, download it, or take the set.
13 skills, 19,760 words, version 1.0.0, written October 2026.
By Afeez Adebayo, Marketing, SEO & Growth Expert with 10+ years of experience.
Created 11 October 2026. Updated 11 October 2026. Free to use under CC BY 4.0.
CRO Audit
cro-audit · audits and research · 1,531 words · start here
Audit a website or app conversion journey end to end: tracking, segmented funnels and sized drop-offs, landing, product, pricing, sign-up and checkout templates, forms, trust and social proof, mobile and speed, qualitative evidence from recordings, surveys and user tests, and findings written as hypotheses, separated into fixes, tests and research and prioritised.
Conversion Research Methods
conversion-research · audits and research · 1,516 words
Plan and run conversion research that produces strong hypotheses: funnel and segment analysis, heatmaps, session recordings, on-page polls, exit and post-purchase surveys, user testing, interviews, support and review mining, form analytics and expert review, with privacy safeguards, triangulation and synthesis into prioritised hypotheses.
Landing Page and Homepage Optimisation
landing-page-optimisation · pages and journeys · 1,514 words
Optimise homepages and landing pages for any traffic source: five-second clarity, visitor intent and message match, page anatomy from headline to closing call to action, one primary action, proof near decisions, objection handling, homepage routing by audience, mobile and speed checks, and prioritised hypotheses and test ideas.
Form Optimisation
form-optimisation · pages and journeys · 1,519 words
Improve sign-up, lead, quote, contact and application forms: field-level analytics, field justification, ordering and multi-step design, labels, input types and autocomplete, flexible formats, inline validation and specific errors, privacy reassurance, mobile and accessibility fixes and changes judged on completion and qualified leads.
Pricing Page Optimisation
pricing-page-optimisation · pages and journeys · 1,545 words
Improve pricing pages so visitors choose a plan with confidence: plan names and one-line descriptions, a highlighted recommended plan, clear prices, currency, tax and billing toggles, value metrics and limits with examples, outcome-grouped comparison tables, calculators for usage pricing, FAQs, enterprise paths, mobile layout and research-backed tests.
Checkout and Cart Optimisation
checkout-optimisation · pages and journeys · 1,514 words
Reduce cart and checkout abandonment: total costs shown early, guest checkout, fewer fields with address lookup and autofill, wallets and local payment methods, delivery dates, trust cues and returns near payment, inline validation, mobile speed, subscription and international details, fixes before tests and consented cart recovery.
Sign-Up Flow and Onboarding Conversion
signup-flow-optimisation · pages and journeys · 1,503 words
Optimise sign-up and early onboarding flows from button click to activation: account creation, social sign-in, SSO and magic links, verification that does not block entry, setup questions that personalise, templates and sample data, card-required, no-card and freemium choices, and tests judged on activated users and paid conversion.
Popups and On-Site Messaging
popups-onsite-messaging · persuasion and messaging · 1,524 words
Plan on-site popups, banners, slide-ins and prompts that help rather than annoy: formats, exit-intent, scroll and page-type triggers, frequency caps and suppression, offers matched to pages, mobile interstitial risk for search, clear consent, accessible design and net effect measured against holdouts on conversion and revenue.
Social Proof and Trust Signals
social-proof-trust · persuasion and messaging · 1,501 words
Use social proof and trust signals where visitors doubt: ratings with counts, reviews filtered by use case, relevant logos, named testimonials, case study numbers, guarantees, policies, real security badges and contact details, placement at points of doubt, credibility improvements, compliant review collection and tests on conversion.
Conversion Copywriting and Value Proposition Tests
conversion-copywriting · persuasion and messaging · 1,527 words
Write and test conversion copy from customer language: voice of customer mining, awareness stages, value proposition angles, page copy ordered by visitor questions, outcome-focused calls to action, reassuring microcopy and helpful errors, clarity editing and tests of different messages judged on conversion and revenue.
Mobile Conversion Optimisation
mobile-conversion · persuasion and messaging · 1,506 words
Close the gap between mobile and desktop conversion: device funnels, real-phone testing, speed and Core Web Vitals by template, first-screen layout, tap targets, navigation, forms with correct keyboards and autofill, wallet payments, pop-up and sticky element control, in-app browsers, cross-device save and share, and mobile tests.
CRO Test Planning and Prioritisation
cro-test-planning · programme · 1,526 words
Build a CRO test roadmap: evidence-based hypotheses, traffic and duration estimates per page, PIE, ICE or PXL scoring, decisions to test, implement, research or drop, low-traffic strategies, scheduling without overlap, test briefs with metrics and decision rules, and a shared learning log.
CRO Programme Reporting
cro-reporting · programme · 1,534 words
Report CRO programmes credibly: conversion by funnel step, device and source, revenue per visitor, average order value and qualified lead rate, test velocity, clear result and win rates, every result including losses and nulls, impact of shipped changes estimated with ranges and overlap discounts, learnings and a roadmap view.
No skill matches that search. Clear it or choose All.
What CRO skills do for an AI agent
Ask a general AI model to “improve conversions” and you usually get a list of best practices: change the button colour, add urgency, put testimonials on the page. Some may help; most are guesses that ignore where your visitors actually struggle and why. Worse, they treat every site the same.
These skills work from evidence. They verify tracking, map the funnel by device and source, size each drop-off and use recordings, surveys and user tests to explain it. Every finding becomes a hypothesis with evidence, a change, an audience and a metric. Obvious bugs get fixed without testing; strong ideas on busy pages get tested; low-traffic pages get bigger changes or more research.
CRO, UX and experimentation
Conversion rate optimisation improves how many visitors take a valuable action. It overlaps with UX research and experimentation but focuses on business outcomes such as revenue per visitor and qualified leads. The skills cover audits and research, pages and journeys, persuasion and messaging, and the programme that ties them together. They work with any analytics, testing and research tools.
Where to start: audit the journey
Start with the cro-audit skill. It checks tracking, maps the funnel, reviews key templates, gathers qualitative evidence and turns findings into prioritised hypotheses, fixes and research.
If you already know where visitors drop off but not why, run conversion-research. If you have many ideas and limited traffic, use cro-test-planning before launching more tests. If carts are abandoned, go straight to checkout-optimisation.
Which skill fixes which problem, and the evidence
Use this table to go straight to the right skill when you already know the symptom.
| Symptom or task | Likely cause | Skill to use |
|---|---|---|
| Conversion fell and nobody knows where | No end-to-end CRO audit | cro-audit |
| Test ideas are guesses | No conversion research | conversion-research |
| High bounce on key pages | Unclear message, weak hierarchy or proof | landing-page-optimisation |
| People abandon forms | Too many fields, poor validation | form-optimisation |
| Visitors leave the pricing page | Unclear plans and costs | pricing-page-optimisation |
| Carts abandoned at checkout | Surprise costs, forced accounts, missing payments | checkout-optimisation |
| Sign-ups never activate | Friction between click and first value | signup-flow-optimisation |
| Popups annoy visitors | Poor timing, targeting or design | popups-onsite-messaging |
| Visitors hesitate at decisions | Proof and trust missing where doubt arises | social-proof-trust |
| Copy does not convert | Company language, unclear value | conversion-copywriting |
| Mobile converts far below desktop | Speed, layout, forms and payments | mobile-conversion |
| Too many ideas, little traffic | No prioritised test roadmap | cro-test-planning |
| Leadership doubts CRO value | Summed lifts, hidden losses | cro-reporting |
The evidence behind these skills
CRO advice is full of best-practice lists that may not apply to your site. These skills rely on research, data and tests, checked on 11 October 2026:
| Finding | Source | Label |
|---|---|---|
| Average documented cart abandonment has long been reported at around 70%, much of it from fixable causes such as surprise costs and forced accounts | Baymard Institute research | Study |
| Intrusive interstitials that block content on mobile can hurt how pages perform in search | Google Search Central | Confirmed |
| Small user tests reveal usability problems but not how often they occur | Nielsen Norman Group | Consensus |
| Adding up individual test lifts overstates programme impact | Experimentation research (Kohavi et al.) | Consensus |
| Fake or misleading reviews are targeted by consumer regulators such as the UK CMA and US FTC | CMA and FTC guidance | Confirmed |
Example prompts for each skill
Each skill triggers on everyday wording. These prompts are good starting points:
- Audit: “Traffic grew but sales did not. Run a CRO audit.”
- Research: “Plan research on why mobile visitors abandon checkout.”
- Landing pages: “Our homepage confuses visitors. Fix the first screen.”
- Forms: “Our quote form has 14 fields. Improve it.”
- Pricing: “Visitors leave our pricing page. Why?”
- Checkout: “Reduce our cart abandonment.”
- Sign-up: “Many people click Start free but never activate.”
- Popups: “Are our popups hurting revenue?”
- Proof: “Where should reviews go on product pages?”
- Copy: “Test three value propositions for this landing page.”
- Mobile: “Mobile converts at half the desktop rate.”
- Planning: “Prioritise our 80 test ideas.”
- Reporting: “Build a CRO report leadership will trust.”
What the agent needs from you
CRO depends on data about real visitors. The most useful inputs are:
- Analytics: funnels and conversion by page, device, source and segment.
- Behaviour data: heatmaps, session recordings and form analytics.
- Voice of customer: surveys, user tests, reviews and support tickets.
- Key URLs: landing, product, pricing, sign-up and checkout pages.
- Traffic levels: weekly visitors and conversions per page for test planning.
- Test history: past results, including losses and nulls.
If something is missing, the skill works with what you have and says which conclusions would change with more data.
What an agent cannot do, and myths to ignore
An agent can analyse, write hypotheses, draft copy and plan tests, but it cannot run tests on your site without access, and it cannot replace watching real users struggle. Decisions about trade-offs, such as removing a lead form field sales wants, stay with you; the skills supply evidence and options.
| Myth | What the evidence says |
|---|---|
| Best practices work everywhere | Visitors and contexts differ; research and tests decide |
| Button colour tests drive growth | Value proposition, friction and trust changes move results more |
| More popups mean more conversions | Measure net effect; intrusive popups can hurt revenue and search |
| Winning test lifts add up | Overlap, novelty and false positives shrink combined impact |
| Every page can be A/B tested | Low-traffic pages need bigger changes or research instead |
Frequently asked questions
What is conversion rate optimisation?
Improving the share of visitors who take a valuable action, such as buying, signing up or requesting a demo, through research, better pages and tests.
Where should CRO start?
With an audit: verify tracking, map the funnel by device and source, find the biggest drop-offs and research why they happen.
How much traffic do I need for A/B tests?
It depends on baseline conversion and the effect you want to detect. Low-traffic pages need bigger changes, longer tests or research instead.
What should I test first?
High-traffic, high-value steps with strong evidence of a problem, such as checkout or pricing pages.
Do popups hurt conversion?
They can. Use relevant, well-timed messages and measure net effect against visitors who see none.
How do I reduce cart abandonment?
Show total costs early, offer guest checkout and familiar payment methods, simplify forms and build trust near payment.
How should CRO results be reported?
By revenue per visitor and conversion by step, with all test results shown and combined impact estimated with ranges.
How do these skills relate to the other marketing skills?
SEO and Performance Marketing bring visitors; CRO turns more of them into customers; Growth and CRM skills keep them.
Sources and more marketing skills
- Baymard Institute UX research
- Nielsen Norman Group articles
- Avoid intrusive interstitials, Google Search Central
- Kohavi, Tang and Xu, Trustworthy Online Controlled Experiments (2020)
- CXL conversion optimisation guides
Checked on 11 October 2026. Pair this set with the Growth Marketing skills for experiment design and the Performance Marketing skills for paid traffic. See every subcategory on the Marketing Skills-md page.
