Customer journey mapping: the branches nobody draws

What a customer journey map is

A customer journey map is a visual model of every stage a customer moves through with a brand, from awareness and consideration through purchase, delivery, service and repeat purchase. Each stage records touchpoints, what the customer is trying to do, what they feel, and where they drop off.

What actually happens after checkout is delivery delays, 'where is my order' tickets, and silent churn, the customer who decides they will just buy somewhere else next time. Most of the reasons customers do not return happen after checkout, which is exactly the stretch most journey maps compress into one optimistic arrow.

It is a genuinely useful planning artifact, and most maps are detailed and well researched up to checkout, then thin afterward. To me that is the biggest blind spot in e-commerce right now: not a new acquisition channel, but what happens after the buy button. Most brands still treat post-purchase as an afterthought, which is wild, because fixing it pays off fast.

Every order is a promise. Keeyu keeps it. The journey map usually treats that promise as a single box labeled delivery.

The ecommerce customer journey, stage by stage

The ecommerce customer journey is the full path a shopper takes with an online store, from first hearing of the brand, through comparing and buying, to receiving the order, getting help with it and buying again. Shopify's guide to ecommerce customer journeys breaks it into five stages: awareness, consideration, acquisition, service and loyalty.

Five stages is a fair summary from the shopper's side. My issue is with what sits between acquisition and loyalty. In that model the only thing after payment is service, which Shopify describes as the point where a shopper needs help with an order. Waiting for the order and receiving it get no stage of their own, so the first post-purchase event on the map is a request for help. I split that stretch into three stages, because each has a different owner and a different signal. That gives seven.

Stages one to three: where shoppers drop off in plain sight

Awareness, consideration and purchase are the pre-purchase stages, and they are where a shopper can leave without buying. Every one of those exits leaves a record. Baymard Institute's average across 50 studies is a 70.22% cart abandonment rate, and leaving out shoppers who were only browsing, 18% of US shoppers who abandoned a cart said the site wanted them to create an account and 10% said their card was declined. To find your own exits, the funnel exploration and path exploration reports in Google Analytics show the step where shoppers dropped out.

My take here is short, because this is not what Keeyu does. I've seen companies chase flashy features while their checkout process frustrated customers into abandoning carts completely, so fix the basics on that list first. The part I care about is the promise. The delivery date and the returns terms a shopper reads in stage two are a commitment that stages four to six have to keep, and no funnel report shows whether they were kept.

Stages four to six: where the promise is kept or broken

Order promise, delivery, and service and returns are the post-purchase stages: everything from payment until the order is complete or refunded. After payment nobody abandons anything, so no funnel report lights up. The shopper has paid and is holding a date, and nothing on a standard map says whether that date is still going to be met. Even with strong tools like Shopify, NetSuite and several 3PLs, the systems are often disconnected across the order journey, which is why a map built from one tool's data stops at checkout and the helpdesk ends up downstream of the issues. For the step-by-step version of these three stages, see the post-purchase customer journey.

Two numbers make these stages measurable:

• Promise-kept rate: of the orders whose promised date fell in the period, the share that arrived complete by that date. If 4,000 orders were due in a month and 3,640 arrived by the date the shopper was given, the rate is 91%, and 360 shoppers had a promise broken. The full formula, and why carrier reports read higher, is on the delivery promise page.
• First detection time: the hours between a break occurring and the team knowing about it. If the first signal is the shopper's ticket, the team found out last. How to instrument it is in first response time vs first detection time.

Many teams cannot produce either number without running reports from several systems, correlating them by hand and investigating order by order, because nothing matches the promise on each order to its real-time fulfillment status. Returns need the same clock, because a return that takes too long drives a ticket. If a customer drops something off and it takes 14 days, they are not happy people.

A support ticket in stages four to six means the shopper found out before the team did. Helpdesk manages complaints. Keeyu prevents them. Keeyu links every order to its promise and tracks it until the order is delivered or the return is refunded. When an order stops tracking to that promise, detection kicks in and a workflow either fixes it automatically or puts it in front of the team. The helpdesk still does its job, and Keeyu integrates with Gorgias and Zendesk. It just should not be the first place a broken promise shows up.

Stage seven is decided earlier

Loyalty is the stage where a shopper buys again, and it carries an outsized share of revenue: in Gorgias's analysis of its own merchants, repeat customers were 21% of customers and generated 44% of revenue. My view is that the second order is won or lost before this stage starts: the loyalty battle is won or lost in the 3-7 days after checkout. Loyalty breaks after checkout, which is why I split repurchase by whether the first order failed.

The decision rule for the whole map: every stage gets an owner, a signal, a time threshold and a remedy. If any of the four is blank, the stage is drawn but not run. If I could change one thing on most maps, it would be the owner column for rows four to six: one team, with customer service at its center, owning the order from payment through delivery and returns. What follows on this page covers rows four to six: the failure paths, who owns them and how to instrument them. Every order is a promise. Keeyu keeps it.

The part of the map that gets compressed

Pre-purchase stages get mapped in detail because that is where marketing owns the work and the analytics are rich. Post-purchase typically collapses into three boxes: order confirmation, shipping notification, delivered.

In reality that stretch contains the highest-emotion moments of the entire relationship. The customer has paid and is now waiting, which means every day of silence is interpreted. And it contains the most branching paths, because this is where things go wrong.

Mapping the failure paths

A post-purchase map that reflects reality includes the branches, not just the happy path:

• Order placed but never synced to the warehouse
• Stock sold that did not exist
• Fulfillment hold nobody actioned
• Shipment created but never collected
• Parcel stalled, delayed or lost in transit
• Delivered to the wrong place or marked delivered without arriving
• Return requested, shipped, received, and not refunded

Each of those is a distinct customer experience with its own emotional arc, and each has a moment where you could have intervened before the customer noticed. That moment is the most valuable thing on the whole map and it is almost never drawn.

If you want the failure paths mapped properly, there are more of them than a workshop will produce. We monitor around eighty points of failure across the post-purchase journey, which I mentioned on eCommerce Australia. A map drawn from customer emotions gives you five or six stages. A map drawn from where orders actually stop gives you a list your operations team can work through on a Monday.

Who owns the post-purchase stretch

The structural problem is ownership. Marketing owns pre-purchase, support owns the complaint, and operations owns the warehouse, so the space between checkout and delivery belongs to nobody in particular.

I learned this at SurfStitch. I did not know a lot about e-commerce when I walked in, but the former CEO and founder had customer service reporting to him. When I moved to P.E Nation, that was the first thing I did: combine customer service and operations into one department, and we saw a dramatic change in the customer experience as a result. The reason is visibility. Everything that happens upstream, through the warehouse and payments, becomes a ticket and becomes customer service's problem, and if customer service is siloed from operations, they only find out when the shopper reaches out.

Ownership decides whether the map ever gets used. The brands doing it properly put customer service and e-commerce operations in one department reporting to the CEO, which is the pattern I described on the Retail Fest post-purchase panel. A journey map drawn by a team that does not own the failure paths becomes an artifact rather than an instrument.

Turning the map into instrumentation

A map is only worth the work if it changes what you measure. For each failure branch, define the signal that says it is happening, the time threshold that makes it actionable, and the remedy.

Then instrument it. Count how many orders take each branch, how many customers found out before you did, and how long each break sits before anyone acts. Expect the busy branch to change. The storefront, payment gateway, ERP and warehouse system are each designed to perform a function, and none of them is designed to connect to the promise the shopper was given at checkout, so when one fails, it fails silently. One week you could have a run of stockouts, the next a batch of fraud holds, a SKU mismatch, a sync issue or orders that were never manifested. A map that only instruments last month's problem will miss this month's.

Instrumentation has a bill attached when you skip it. Around 123 million hours went into customer service calls in a single year, a figure I cited on Add To Cart. The biggest post-purchase mistake I see is not communicating with the shopper when something has gone wrong, and waiting for them to reach out instead. A map with no signal on its failure branches makes that the default, because the shopper's message is the first anyone hears of the break.

Where this sits

This is proactive e-commerce operations: give the post-checkout journey an owner and a detection layer, so the failure branches get caught before the customer walks down them. Detect. Decide. Act. Keeyu gets shoppers what they want, on time, as promised. Every order is a promise. Keeyu keeps it.

See how the workflows run or book a demo.

Related reading

For the experience layer, read customer experience management. For metrics, see customer satisfaction metrics. For the operational picture, read post-purchase operations.

• Customer retention strategies: fix the leak before the campaign
• Proactive vs reactive customer service: the real difference
• Returns management, and the returns that were never really returns

• Harvard Business Review, 2017: 73% of 46,000 studied shoppers used more than one channel.
• Baymard Institute, cart abandonment meta-study: average cart abandonment at 70.22% across 50 studies, and the reasons US shoppers give, including forced account creation (18%) and a declined card (10%).
• American Customer Satisfaction Index, retail study 2025: where online retail satisfaction actually moves.
• The Breakout CEO #86, The Pivot This Founder Made After an Investor Called It Impossible - one in five US shoppers not getting orders on time, and the ticket ratios behind it.
• The 9-5 Exit Plan, How AI Is Making Customer Support Invisible - proactive AI and support that disappears.
• Retail Fest 2026, Post-Purchase Perfection: Delivering Experiences Customers Rave About - why a help desk cannot prevent a ticket.
• Shopify Enterprise, Ecommerce Customer Journeys: Stages and Mapping: the five-stage model (awareness, consideration, acquisition, service, loyalty) and Shopify's description of the service stage.
• Gorgias, The CX-Driven Growth Playbook, chapter 1: in Gorgias customer data, repeat customers are 21% of customers and generate 44% of revenue and 46% of orders.
• Google Analytics Help, Funnel exploration: the report that shows how users succeed or fail at each step of a path.
• Google Analytics Help, Path exploration: the report that traces the steps users take, used here to find where shoppers leave before buying.