Part one · The journey you can measure · Chapter 5

WHERE CUSTOMERS WAIT

Conversion says how many customers move on. Time in stage says when, and how long the slow ones wait while you say nothing useful.

Two stages can have the same conversion rate and very different journeys. In one, customers move on in a tight window. In the other, half move quickly and the rest drift for weeks, and that’s where flows fire at the wrong moment.

Median and 75th percentile, not the average

For each stage, take the customers who moved on to the next one and measure how many days it took. Report two numbers: the median (half were faster) and the 75th percentile (three in four were faster). Skip the average, which a few very slow customers drag far from typical. The gap between the two is the stage’s tail.

The target window

Then set a target for each stage: the day by which a customer should have moved on if the journey is working. It can come from a promise (delivery in five days), the product (a 30-day supply) or judgment (first use within a week). Write it down before you look at the data.

The stall in time is the stage where the 75th percentile runs furthest past the target. That’s where the slow quarter of your movers, and everyone behind them who never moves, wait longest after they should have acted.

DerivedThe number on the right is the 75th percentile divided by the target. These are the defaults in the tool below.

In days, the reorders take longest. Against target, first use is worse: the slow quarter take 2.6 times as long as they should, in a stage that often gets no useful message at all.

Silence means different things for different customers

The repeat-purchase models marketers use to predict which customers are still active, such as Peter Fader, Bruce Hardie and Ka Lok Lee’s BG/NBD model, need only two facts per customer: recency, when they last bought, and frequency, how often Published. In those models, thirty quiet days means more for a customer who ordered every two weeks than for one who orders twice a year. Judge time in stage the same way: against what’s normal for that stage and that kind of customer. The Second Order covers timing the reorder reminder; this chapter is about finding which stage to time first.

Run your numbers

Which stage keeps customers waiting longest past its target?

Example numbers. Replace with yours. Days in stage are for customers who moved on to the next stage, measured from the day they entered it.
Order to delivered
Delivered to first use
First use to second order
Second order to third order
the stall in time
75th percentile as a multiple of the target
days the slow quarter waits past the target
tail: 75th percentile over the median
Only customers who moved on are timed. A tail above about 2× usually means two groups in one stage: split it by product, channel or discount first.

With the defaults, first use is the stall in time: its 75th percentile is 2.6 times its target, and the slow quarter waits 11 days past the 7-day target, with a tail of 3.0 (the 75th percentile is three times the median). The median is on time, so this is a fast group and a slow group, not a bad target.

Do this

If this is your store

See where your first-time buyers stall.

The free Klaviyo audit reads your own account and scores it. Start with your store’s address. A read-only key gets the full report.

Get the free audit →

Written by Andrew Lauchner, a growth and retention operator for consumer brands. The paid work is one ninety-day Sprint.

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.