Ecommerce Customer Journey Analytics: How to Find Conversion Problems
How to analyse real shopper journeys, not just funnels: path analysis, entry points, loops, multi-session journeys and qualitative evidence to find what's broken.
Quick answer
Customer journey analytics finds conversion problems by studying the paths shoppers actually take rather than an idealized funnel. Start from entry points (landing pages by channel), follow the next pages and actions with path analysis, look for loops, detours and exits after policy or shipping pages, compare paths that end in purchase with those that don't, and measure multi-session behavior such as days and sessions to purchase. Treat each pattern as a question, confirm the cause with recordings, surveys or user tests, then fix and verify.
Why Funnels Aren't Enough
Funnels are essential for measuring drop-off between fixed steps; see ecommerce conversion funnel. But shoppers don't move in straight lines. They land on a product from an ad, go back to a category, search, check returns, leave and come back through email two days later. The diagram above shows how varied entry points and routes are, and where exits cluster.
Journey analytics complements two other methods: qualitative journey mapping, covered in the Shopify customer journey audit, and the platform-level funnel.
Journey Questions Worth Answering
| Question | Analysis |
|---|---|
| Where do buyers enter, and where do non-buyers enter? | Landing pages by channel with conversion and next page |
| What do shoppers do after landing? | Forward path exploration from key landing pages |
| What happens before shoppers leave? | Backward paths from exits on product and cart pages |
| Do shoppers loop? | Repeated sequences between category, product and search |
| Which paths lead to purchase? | Compare paths of converting and non-converting sessions |
| How long do journeys take? | Days and sessions to purchase by category and channel |
Step 1: Start From Entry Points
Group sessions by landing page type and channel. A shopper landing on a product page from paid social has different needs from one landing on the homepage from brand search. For each group, look at bounce or engagement, the next page viewed and conversion. Entry points that send people straight back to the homepage or to search often mean the landing page didn't match what the ad or search result promised.
Step 2: Follow Paths Forward and Backward
GA4's path exploration shows the sequences of pages or events that follow a starting point, or precede an ending point. Use forward paths from your top landing pages and backward paths from exits on product, cart and checkout pages. Keep path depth short; after three or four steps, patterns fragment.
Step 3: Look for Telltale Patterns
| Pattern | Possible meaning |
|---|---|
| Category ↔ product back-and-forth | Listing cards lack the information needed to choose |
| Search → refine → search → exit | Search results don't match intent |
| Product → shipping or returns page → exit | Costs or policies are a deal-breaker or unclear |
| Cart → policy page → exit | Last-minute trust or cost concern |
| Checkout → back to cart → exit | Unexpected cost or forced account at checkout |
| Product → size guide → exit | Sizing information isn't answering the question |
Worth noting
Patterns are hypotheses, not conclusions. A visit to the returns page may be diligence from a buyer, not a problem. Compare with converting sessions before acting.
Step 4: Compare Buyers and Non-Buyers
The most useful comparison is between sessions that purchased and similar sessions that didn't. Do buyers use search more? View more images? Open the size guide? Visit fewer pages before adding to cart? Differences point to what helps people decide and where non-buyers get stuck.
Know where shoppers leave, but not why?
ZSpace combines journey data, recordings and user research to find the cause behind your store's drop-offs.
Step 5: Measure Multi-Session Journeys
Many purchases, especially higher-priced ones, take several visits. Measure days to purchase and sessions to purchase by category and channel, and look at which channels bring shoppers back (email, retargeting, brand search). Identification is the limit: without logins, a shopper on phone and laptop looks like two people, so cross-device journeys are only partly visible.
Step 6: Add the Why
Quantitative paths show where. For why, watch session recordings filtered to the pattern you found, run a short on-site survey on the page where people leave, read support tickets and reviews, and run moderated user tests on the path. See ecommerce heatmaps and usability testing.
Step 7: Fix and Verify
Turn each confirmed cause into a change with a measurable expected effect on the journey, for example fewer category–product loops and a higher add-to-cart rate from listing pages. Test where traffic allows, and recheck the same path analysis after release. See ecommerce experimentation framework.
Set Up for Journey Analysis
- Consistent page naming or content grouping by template type
- Standard ecommerce events plus key interactions: search, filter, size guide, shipping info
- Channel and campaign tagging that survives redirects
- Checkout tracked without breaking sessions
- Logged-in customer IDs sent where consent allows
- Recording and survey tools that can be filtered to paths
Journey Maps vs Journey Analytics
Journey maps, built in workshops and research, describe what customers are trying to do, their questions and feelings at each stage. Journey analytics measures what they actually do on your site. Use both: the map suggests where to look and what questions to ask; analytics shows where the map is wrong and how often each path happens. When they disagree, investigate with qualitative research. See conversion research.
Cross-Device and Consent Limits
Journey analytics sees only what it can connect. Shoppers who browse on a phone and buy on a laptop appear as two journeys unless they sign in. Visitors who decline analytics consent may be missing or modelled. Ad blockers and short cookie lifetimes break long journeys. Treat multi-session analysis as a partial view, prefer signed-in data for long journeys, and avoid over-interpreting precise path frequencies. See ecommerce attribution for how the same limits affect channel credit.
Journey Analysis Toolkit
| Question | Method |
|---|---|
| Where do journeys start? | Landing page and source reports |
| What do buyers do that non-buyers don't? | Path comparison, event sequences |
| Where do people loop? | Path exploration, repeated page views |
| How long do journeys take? | Time and sessions to purchase |
| Why do they stall? | Recordings, surveys, usability tests |
| Which customers return and buy? | Cohort and retention analysis |
Common Mistakes
- Treating a single dominant path as the whole story
- Reading patterns without comparing to converting sessions
- Paths too deep to interpret
- Assuming cross-device journeys are complete
- Fixing before confirming the cause
Want to see how shoppers really move through your store?
Talk to ZSpace about a conversion audit and UX research.
Conclusion
Journey analytics shows the routes shoppers take and where those routes break. Start from entry points, follow paths, look for loops and exits, compare buyers with non-buyers, measure multi-session behavior, and confirm causes before fixing. It turns a vague “conversion is low” into specific, testable problems.
For related guides, see product analytics and event tracking.
Common questions
Analysing the actual sequences of pages and actions shoppers take, across one or several sessions, to see which paths lead to purchase and where shoppers loop, stall or leave.