A/B testing: what it is, how to do it right, and mistakes that ruin your tests
A/B testing compares two versions of an element to determine which converts better. Learn the correct methodology, common mistakes, and how to scale your...

The customer journey is the complete sequence of interactions a person has with your brand from the moment they identify a need until they become a returning customer. It's not a linear funnel — it's a map of decisions, emotions, and friction points.
According to McKinsey data, companies that actively optimize their customer journey increase revenue by 10% to 15%, while reducing service costs by 15% to 20%. This isn't theory — it's a direct impact on the bottom line.
The problem is that most teams work in silos. Marketing optimizes acquisition, product improves the interface, and support handles complaints. Nobody looks at the full journey. The result: fragmented experiences that lose customers in the transitions between stages.
If you want to dive deeper into how the overall experience affects retention, check out our complete guide to customer experience.
There are multiple models, but the most practical one for digital optimization divides the journey into five clear stages:
The user identifies a problem or need. They don't know your brand yet and aren't aware that a specific solution exists. They search on Google, see an ad, or a colleague mentions a product category.
Key metrics: impressions, reach, organic traffic from informational keywords, ad CTR.
Typical friction: generic content that doesn't connect with the user's real pain point.
The user already understands their problem and is evaluating options. They compare providers, read reviews, and download resources. This is where your value proposition is critical.
Key metrics: time on page, pages per session, lead magnet downloads, newsletter subscriptions.
Typical friction: lack of social proof or clear differentiation from competitors.
The moment of truth. The user is ready to buy, but any friction can kill it: a confusing checkout, unexpected costs, lack of trust signals.
Key metrics: conversion rate, cart abandonment rate, average order value (AOV).
Typical friction: long forms, limited payment options, lack of visible guarantees.
Post-purchase. The customer uses your product or service. Their experience determines whether they return, recommend, or leave.
Key metrics: NPS, repeat purchase rate, support tickets, churn rate.
Typical friction: poor onboarding, slow support, unmet expectations versus what was promised in marketing.
The satisfied customer becomes an ambassador. They recommend, leave positive reviews, and generate new organic traffic.
Key metrics: referrals, reviews, social media mentions, UGC (user-generated content).
Typical friction: not facilitating or incentivizing recommendations.
A customer journey map isn't a pretty diagram to hang on the wall. It's an operational tool that identifies where you're losing money.
Step 1: Define your buyer personas. Not generic personas — people with real context, motivations, and objections. Without this, the map will be fiction.
Step 2: Identify the touchpoints. List every point of contact: ads, landing pages, emails, checkout, support, packaging. Use real analytics data, not assumptions.
Step 3: Map the emotions. At each touchpoint, how does the user feel? Trust? Confusion? Frustration? Session recordings (Hotjar, Clarity) are gold for this.
Step 4: Detect the moments of truth. These are the 3-5 points where the user decides to continue or abandon. They are your optimization priorities.
Step 5: Assign metrics to each stage. Without data, there's no optimization. Each stage needs specific KPIs.
Step 6: Prioritize by impact. Don't try to fix everything at once. Prioritize the friction points with the highest traffic volume and the biggest conversion drop.
This practical framework helps you diagnose where your journey is failing:
| Stage | Friction signal | Diagnostic tool | Typical action |
|---|---|---|---|
| Awareness | Bounce rate >70% on landing pages | GA4 + heatmaps | Review ad-to-landing page match |
| Consideration | Time on page <30s on comparison pages | Scroll maps + surveys | Improve content structure |
| Decision | Cart abandonment >75% | Funnel analysis + session recordings | Simplify checkout, add trust signals |
| Retention | Churn >5% monthly | NPS surveys + ticket analysis | Improve onboarding and post-sale communication |
| Advocacy | <2% of customers leave a review | CRM + email automation | Referral program, post-purchase email |
For a more granular analysis with real data, check out our customer journey analytics guide.
Mistake 1: Mapping the journey you want, not the one that exists. 80% of journey maps I see in audits are aspirational. They reflect what the team believes is happening, not what the data shows.
Mistake 2: Ignoring micro-moments. Between the major stages, there are dozens of micro-interactions that accumulate friction. A button that doesn't stand out, ambiguous copy, a 4-second load time. These are the silent killers of conversion.
Mistake 3: Not segmenting by user type. A new user and a returning user have radically different journeys. Treating them the same wastes opportunities.
Mistake 4: Creating the map and never updating it. The journey changes with every product launch, every market shift, every new feature. A map from 6 months ago is already obsolete.
Mistake 5: Not linking the map to concrete actions. A journey map without a prioritized optimization backlog is an academic exercise.
Design thinking provides a structured methodology for improving the journey:
Empathize: Interviews with real customers (at least 5-8). Not mass surveys — 30-minute conversations where the user narrates their experience.
Define: Synthesize findings into "Jobs to be Done" (tasks the customer needs to complete at each stage).
Ideate: Generate improvement hypotheses for the 3-5 highest-friction points. Use both quantitative and qualitative data.
Prototype: Before developing, test with prototypes. A wireframe in Figma costs 2 hours; a poorly focused development costs weeks.
Test: Validate with A/B testing. Every journey hypothesis should translate into a measurable experiment.
For mapping: Miro, Whimsical, or even a well-structured spreadsheet. The tool matters less than the discipline of using real data.
For diagnosing: Google Analytics 4 (funnels and path exploration), Hotjar/Clarity (recordings and heatmaps), on-site surveys (Hotjar Surveys, Qualaroo).
For optimizing: A/B testing tools (VWO, AB Tasty, Optimizely) to validate each change before deploying it to production.
For automating: CDP (Customer Data Platform) that unifies data from all touchpoints and enables real-time personalization.
A fashion ecommerce site with 200K visits/month found that its conversion rate was at 1.2% — below the industry benchmark (1.8-2.5%).
Diagnosis by stage:
Actions implemented:
Result: conversion rate went from 1.2% to 1.9% in 8 weeks. AOV increased by 12% thanks to the free shipping bar.
You don't need 50 KPIs. You need the right ones at each stage:
| Stage | Primary KPI | Indicative benchmark |
|---|---|---|
| Awareness | Organic / paid CTR | 3-5% organic, 2-4% paid |
| Consideration | Engagement rate (GA4) | >55% |
| Decision | Conversion rate | 1.5-3% ecommerce, 2-5% SaaS |
| Retention | Repeat purchase rate / Churn | >25% repeat purchase within 90 days |
| Advocacy | NPS | >50 excellent, >30 good |
The customer journey is not a one-day project. It's an ongoing process of measurement, hypothesis, and validation. Start by:
If you need professional help mapping and optimizing your customer journey, at Boost we specialize in CRO. You can also start with a free diagnostic of your website at Scan&Boost.
Adrià Vidal is the founder of Boost. +1,000 optimization actions, +47.8% average conversion uplift per client, +€7.8M in additional revenue generated.
A/B testing compares two versions of an element to determine which converts better. Learn the correct methodology, common mistakes, and how to scale your...
Learn what process automation is, its real benefits, and how to implement it step by step to reduce costs and scale your business.
Discover what a digital strategy is, how to build one step by step, and the 7 pillars every company needs to master in 2026 to grow online.