Stages Are Events

The first decision is the one most maps get wrong: what the stages are. Get it right and the rest of this book is arithmetic.

Four rules for a stage

It starts with an event.Something that happens to one customer at one moment and leaves a record: an order, a scan, a login, a ticket. Not a feeling and not a phase.
Most repeat buyers pass through it.A stage only a few reach is a detour.
Someone can change it.If nothing you control moves it, it’s context.
It nests.Count a customer at a stage only if they also passed every stage before it. That keeps the counts falling in order, and it tests your signals: if many customers reorder without ever showing your first-use signal, the signal is too weak.

Six to nine stages is enough.

The measured journey. The two filled stages are where the example cohort in chapter 4 loses the most customers.

A standard set, and where each event lives

Stage	Event	Where the data is
Arrive	First session with a known visitor ID	Analytics; joined to the customer at signup or checkout
First order	First paid order per customer	`orders`, the earliest non-canceled order
Delivered	Carrier’s delivered scan	`fulfillments` or your tracking app’s events
First use	Your chosen first-use signal	See below
Second order	Second paid order, excluding exchanges and replacements	`orders`, with exchange orders filtered out
Third order	Third paid order, same rules	`orders`
Contact	First ticket, chat or call	`support_tickets`, matched by email or order
Return	Return requested	Returns app export or `refunds`
Lapse	Days since last order pass your 75th-percentile gap	Derived from `orders` (chapter 5)

Appendix A has the query that turns these into one row per customer with a timestamp for each stage.

Finding a first-use signal

First use is the stage that order data can’t see, and it tends to sit in a long silent stretch. You won’t find a perfect signal. You need one that’s good enough, per customer, and cheap to collect. Candidates, by what you sell:

Anything with instructions: a view of the how-to page or video, or a click on the how-to link in the delivery email.
Consumables and supplements: a reply to a day-seven check-in (the template is in Appendix B), a scanned QR code on the pack, a refill reminder opted into.
Beauty and apparel: a review, a photo, a size or shade question after delivery, or the return window passing with no return.
Anything with an app or account: the first login or setup after delivery.

Then test it: customers with the signal should reorder at a clearly higher rate than those without. If not, it measures something else, such as who opens email. Used, Not Bought covers how to design the first-use stage itself.

Name stages for the customer’s progress

Clayton Christensen and colleagues define a job to be done as “the progress that a person is trying to make in a particular circumstance” Published. That’s a useful test for stage names. “Delivered” is your milestone; the customer’s is “I have it and I know how to start”. Where you count only your milestone, add a signal for theirs.

Do this

Write your stages as a list of events, each with the table and field it comes from. Pick one first-use signal and test it against reorder rate this week. If you can’t find one, the check-in email in Appendix B creates one.

This is one chapter of The Journey Map, which is free and readable in full on a single page with no form in front of it.