← Back to work
Case study

Customer journey diagnosis

Combining behavioural evidence, search data and business context to identify where customers were becoming confused, hesitating or leaving.

The situation

A large digital experience served customers with very different needs, levels of knowledge and reasons for visiting. Pages were reviewed and improved one at a time, and judged on their own numbers. A page with a high exit rate could be treated as the problem even when the real cause sat earlier in the journey.

What was missing

The business could see where customers were leaving, but not why. The evidence existed; it just never met. Analytics showed exits and routes. Search data showed the language customers used. Recordings showed clicks, scrolling and hesitation. The commercial side knew about availability, campaign priorities and shifts in demand. Each sat in its own tool and its own report.

Read separately, each source told part of the story, and that part could mislead. A page could look like it was failing when customers had arrived through a misleading route. Repeated clicking could mean confusion, or important information that wasn't visible at the right moment. Without the connected picture, improvement work kept responding to symptoms.

My role

I approached it as a diagnosis rather than a redesign. I reviewed journeys across pages and entry points using recordings and heatmaps, then checked what I saw against search behaviour, page performance and the commercial picture. The job was to separate usability problems from problems caused by content, navigation, availability or expectations set earlier, and to turn a large number of observations into a short list of priorities.

What got written down

I turned the approach into a structured method, so the next journey could be assessed the same way: four lenses applied together, a rule for what counts as evidence, and a way to rank what the evidence finds.

The method the diagnosis ran on

  • Customer intent: what customers were trying to achieve, read from the language they searched with and the routes they arrived by.
  • Observed behaviour: what they actually did. A single recording is never proof; only patterns repeated across customers and sessions count.
  • Journey continuity: every page judged in the context of the steps before and after it, checking that language, expectations and options stay consistent.
  • Business context: availability, campaign activity and content quality, so not every performance problem is treated as an interface problem.
  • Findings are grouped by cause and ranked by likely customer impact, commercial significance and difficulty to resolve.

Section 1

Multiple sources of evidence

Collected independently

Behavioural evidence

  • Session recordings
  • Heatmaps
  • Event tracking
  • Form analytics

Search data

  • Search terms
  • No results reports
  • Abandoned searches
  • Keyword intent

Page performance

  • Page views
  • Exits & drop-offs
  • Timing & speed
  • Conversion rates

Business context

  • Availability & stock
  • Campaigns & priorities
  • Seasonality & demand
  • Content quality
Synthesised into a unified diagnosis

Section 2

Unified diagnosis

Understanding the complete journey

  1. 1 Customer intent What customers are trying to achieve
  2. 2 Observed behaviour What customers actually do
  3. 3 Journey continuity How each step connects before and after
  4. 4 Business context Why behaviour may be happening

Section 3

Where friction appears

  1. Expectation mismatch What was promised earlier doesn't match what's found here
  2. Confusing interactions Customers repeat, retrace or misread how something works
  3. Disconnected journey A step makes sense alone but breaks the flow around it
  4. Limited or irrelevant options What's on offer doesn't fit what the customer needs
  5. Uncertainty at decision Customers hesitate rather than commit at the critical moment

Section 4

How evidence becomes insight

Patterns over noise

Look for recurring signals across sessions and users, not isolated events

Diagnose the real cause

Link evidence to the relevant journey step

Prioritise with impact

Focus on the issues with the greatest customer and business impact

A small set of high-impact opportunities

Clear, evidence-backed, and aligned to the business

Better decisions Evidence-led
Right problems, right solutions Fix the cause, not the symptom
More impact Focus effort where it drives the biggest improvement
Stronger experiences Clearer journeys that build trust

What changed

The business could tell genuine customer friction apart from poor performance caused somewhere outside the page. Problems were prioritised by likely impact rather than visibility, and redesign work started from an accurate picture of what needed to change. Evidence that had lived in separate reports and tools was reviewed together, as a repeatable way to assess future journeys.

The same instinct, treating a journey as one connected system rather than a set of pages, shaped Helping customers find the right option and The final step to conversion, further along the same journeys.

The conversation moved from opinions about individual screens to evidence about the whole journey.

---
title: Customer journey diagnosis
slug: customer-journey-diagnosis
updated: 2026-10-03
summary: Combining behavioural evidence, search data and business context to identify
  where customers were becoming confused, hesitating or leaving.
---

If a problem like this sounds familiar, the audit is where I'd start: one workflow, one 90-minute session, and a written brief on what's missing and what to fix first.

Book the audit

Prefer to talk first? Get in touch — no obligation.