AI PROJECT · JOURNEY ANALYSIS CONCEPT
Turn an Unclear Customer Journey
into Points You Can See and Fix
A concept demonstration of using AI to hypothesise where and why customers stop progressing in the journey, with every hypothesis checked against real context before it's used.
Project status
Concept Prototype
Business problem
Businesses usually know revenue fell. They rarely know at which stage customers stop progressing, or why.
Why the problem matters
Fixing the end of the journey without knowing the drop-off point spends resources in the wrong place and changes nothing.
Who this is for
Built for executives and teams who want a concrete example of structuring a customer journey into measurable stages, with hypotheses that are checked before use, before considering how the approach might apply to their own work.
Project purpose
Provide a way to structure the customer journey into measurable stages, then use AI to hypothesise what causes each drop-off.
Chanikan's role
Designed the stage structure, the metrics for each stage of the journey, and the diagnostic questions for each point where customers stop progressing.
Tools and technology
- A large language model for hypothesis generation
- Google Sheets for fictional journey data
- A self-designed journey map
Workflow overview
Workflow
- Break the customer journey into stages that can be tracked and measured
- Analyse progression and drop-off rates at each stage
- AI suggests possible factors causing customers to stop progressing at the stage with the highest drop-off
- A human tests those against real operating context
- Sequence remediation experiments by impact and effort
Expected output
- A customer journey map showing progression rates and where customers stop at each stage
- A ranked set of cause hypotheses
- A prioritised experiment list
Example application
Illustrative workflow example
Analysing where customers stop progressing in a sample signup journey
Input
- Sample customer journey data
- Progression rate at each stage
- General behaviour data
AI-assisted process
- Analyse progression and drop-off rates at each stage
- Suggest possible factors causing customers to stop progressing
- Rank issues by evidence-based impact
Human Review
- Check hypotheses against real operating context
- Confirm nothing references data that doesn't exist
- Screen out conclusions without supporting evidence
Output
- A customer journey map with the issues to address first
Review by the accountable owner
AI hypotheses are treated as unverified assumptions. They must be checked against real operating context and additional data before informing a decision.
Responsible AI and data boundaries
- Examples use public, sample, or synthetic data
- Customer data, personal data, or confidential data should never be entered into a public demo
- AI output must be reviewed before it is used
- This project demonstrates an approach — it is not a fully autonomous enterprise system
Current limitations
- A concept prototype for demonstrating the analysis approach, not connected to a live system
- Tested only on fictional journey data
- No integration with any analytics platform or CRM
Next development stage
Broaden the fictional dataset across more business models and package it as a course canvas.
Chanikan's methodology
Chanikan Kanchanasalee
AI-Native Business Strategist specialising in Commercial Growth
Experienced in leading cross-functional work across Commercial Growth, Marketing, Sales, CRM, Data and Operations.