Conversion Optimisation
Help people complete a clear, valuable action by removing friction and supporting a confident choiceâwithout deceptive pressure.
Prerequisite: Customer Discovery â know who this is for and what state they arrive in. Related concepts: Feeling North Star, Tools, Techniques, and Practices.
What it is
Section titled âWhat it isâConversion optimisation is the general discipline of helping people complete a specific, valuable actionâcheckout, signup, an application, a bookingâby removing genuine friction and supporting a confident decision, as distinct from Monetisation, which is specifically about the money-changing-hands moment, and Onboarding, which is specifically about a new userâs first session. Any funnel with a defined completion event and a population that drops off along the way is conversion optimisationâs territory.
The discipline has an unusually short distance to its own corruption: the same funnel-analytics toolkit used to find genuine confusion (an unclear price, a broken form, an unnecessary field) is just as easily used to find the point of maximum leverage for pressureâa countdown timer, a pre-checked upsell, a confirmshaming exit link. This handbook treats those as different activities entirely. Legitimate conversion work removes obstacles between an already-motivated person and their goal; illegitimate conversion work manufactures obstacles or false urgency to extract a decision the person would not otherwise make.
Why it works
Section titled âWhy it worksâConversion rates respond to clarity and trust far more reliably, and far more durably, than they respond to pressure. A confusing step, an unexpected cost revealed late, or an unclear next action all produce the same failure modeâhesitation that curdles into abandonmentâand each is fixable with the same tool: making the true cost, time, and outcome of the action legible before the person commits to it. This is why step-level drop-off analysis is more useful than aggregate conversion rate alone: it names the exact screen where confidence broke, rather than leaving the team to guess.
Pressure tactics can lift short-term conversion numbers while quietly destroying the metric that actually matters: post-conversion quality. A checkout that spikes completions through urgency or hidden costs will also spike refunds, chargebacks, and immediate cancellations, because the person converted before they were actually convinced. User Agency is the principle that keeps this strategy honest: a genuine limited offer with a real deadline and a respectful alternative is a legitimate tool; a fabricated one is deferred churn wearing a growth chart.
Core Tools, Techniques, and Practices (TTPs)
Section titled âCore Tools, Techniques, and Practices (TTPs)â- The Paywall: Present a clear, fair upgrade decision when users understand the value on offer
- Limited Offer: Present a genuine, time-bound offer when it creates a fair choice, not pressure
- Intentional Friction: Add a small, purposeful pause when moving quickly could create harm, regret, or an avoidable mistake
- Success Moments: Mark meaningful progress with feedback that feels earned, then let users decide what comes next
- Spark Curiosity: Offer a truthful glimpse of a useful outcome, then let users choose whether to explore it
- Momentum Bias: Give people a genuine head start without taking the decision away from them
Supporting Tools, Techniques, and Practices (TTPs)
Section titled âSupporting Tools, Techniques, and Practices (TTPs)â- Time to Value: Help people reach their first meaningful outcome with the least unnecessary effort
- JTBD Copywriting: Write from the progress the user is trying to make, not the feature your product wants to describe
- Perceived Effort Delay: When work takes time, show the real work; when it does not, respect the userâs time
- Personalisation: Make the product fit the userâs stated needsâwithout making assumptions they cannot inspect or change
Insights & Metrics
Section titled âInsights & MetricsâOptimise for informed completionânot for forcing every visitor through the same call to action (CTA).
- Conversion Rate
- Formula: (Users who complete the defined conversion event á Eligible users in the funnel) à 100
- What it measures: Share of people who take the action you care about
- Why it matters: Core signal of funnel effectivenessâtrack relative to design changes and traffic quality, not in isolation
- Step Drop-off Rate
- Formula: (Users who leave at step N á Users who reached step N) à 100
- What it measures: Where the path loses people
- Why it matters: Names the concrete moment of confusion, doubt, cost surprise, or premature ask
- Median Time to Convert
- Formula: Median time from first eligible touchpoint (or funnel entry) to conversion
- What it measures: How long the decision path typically takes
- Why it matters: Guides nurture windows and follow-up timing; median beats mean when a few long journeys skew the average
- Conversion Rate by Traffic Source
- Formula: (Conversions from source S á Visits or eligible users from source S) à 100
- What it measures: Channel quality and intent fit
- Why it matters: Separates acquisition problems from product/UX problems before you âfixâ the wrong step
- Early Post-Conversion Quality
- Formula: (Refunds, chargebacks, or cancels within 7 days of conversion á Conversions in the same period) à 100
- What it measures: Whether converted users stay willing
- Why it matters: A lift that spikes regret is not optimisationâit is deferred churn
Behind the Data
Section titled âBehind the DataâFind where users lose clarity, confidence, or momentumâand whether âurgencyâ is real or manufactured.
Funnel & friction
Section titled âFunnel & frictionâ- Which step has the highest drop-offâand what ask sits there (account, payment, permission, long form)?
- How do mobile vs desktop (or web vs app) conversion and drop-off compare on the same steps?
- Does shortening the path raise completion and post-conversion qualityâor only the former?
Traffic & segment fit
Section titled âTraffic & segment fitâ- Which sources and campaigns convert bestâand which inflate visits while lowering rate?
- How do conversion and early cancel rates vary by segment, plan intent, or prior product use?
- Are you optimising a step that low-intent traffic can never pass?
Decision quality & timing
Section titled âDecision quality & timingâ- Where do people hesitate longestâand is that hesitation useful (comparison) or broken (unclear price, trust, next step)?
- If you use a Limited Offer, is the deadline genuine and the alternative still a respectful choice?
- What follow-up timing helps abandoned journeys without nagging or shame?
Further reading
Section titled âFurther readingâ- Conversion Rate (Nielsen Norman Group) â Defines conversion rate for UX/analytics and why you should optimise, not blindly maximise it.
- Define Stronger A/B Test Variations Through UX Research (NN/g) â Why experiments need research-backed hypotheses, not guesswork variants.
- Deceptive Patterns in UX (NN/g) â How conversion pressure turns into unethical (and often unlawful) patterns.
- Deceptive Patterns (deceptive.design) â Canonical taxonomy of manipulative tactics, including fake scarcity and fake urgency.
- E-Commerce Cart & Checkout Usability Research (Baymard Institute) â Evidence-led checkout friction patterns that commonly kill completion.
Agent skill
Section titled âAgent skillâ- Primary command / sequence:
/productfeeling sequenceâmap â friction â trust â tone â anti-patterns â review - Core TTPs to load: The Paywall, Limited Offer, Intentional Friction, Success Moments, Spark Curiosity, Momentum Bias
- Supporting TTPs: Time to Value, JTBD Copywriting, Perceived Effort Delay, Personalisation
- When the agent should use this strategy: âcheckout drop-offâ, âsignup conversionâ, âpaywall conversionâ, âfunnel frictionâ, âCTA not convertingâ, âstep drop-offâ
- Companion handoff: Impeccable â form, checkout, and CTA craft on named funnel steps; DocSlime â when conversion definitions or funnel success criteria belong in
docs/experience/; RedTeam â before A/B variants that add urgency, hidden costs, or confirmshaming (/redteam assumptionson whether the lift trades away post-conversion quality) - Feeling north star this strategy serves: clarity and confidence through fair, earned completion
- Anti-goals: fake scarcity/urgency, confirmshaming, hidden costs, forced continuity, A/B tests that only maximise click-through at the userâs expense
- Reference path:
skill/reference/sequence.md