Onboarding Funnel Sankey Diagram: Editable Template
See which onboarding steps new accounts reach and where they stop, and compare accounts that invite a teammate with those that stay solo. Replace the counts with your own.
An onboarding Sankey shows new accounts moving through the first steps of a product and splits them by a behaviour that may matter, such as inviting a teammate. The comparison between the branches is the point.
Each new account either verifies its email or not, creates a first project or not, and then either invites a teammate or works alone. Both of those last branches end in the same two outcomes, weekly active or lapsed, so you can read the difference in outcome directly off the ribbons.
The numbers are invented. A Sankey shows an association in your data, not a cause: accounts that invite teammates may be the ones that were going to stay anyway.
How to read this diagram
- 1,000 accounts are created; 290 (29%) are weekly active at day 30.
- 82% verify their email and 68% of those create a first project.
- Of the 240 accounts that invite a teammate, 180 (75%) are weekly active; of the 320 that stay solo, 110 (34%).
- That gap is a lead to test, not proof that invitations cause retention.
The data behind it
Invented figures for illustration. Not data about any real business.
| From | To | Amount (accounts) |
|---|---|---|
| Accounts created | Verified email | 820 |
| Accounts created | Never verified | 180 |
| Verified email | Created first project | 560 |
| Verified email | Did not start a project | 260 |
| Created first project | Invited a teammate | 240 |
| Created first project | Stayed solo | 320 |
| Invited a teammate | Weekly active at day 30 | 180 |
| Invited a teammate | Lapsed | 60 |
| Stayed solo | Weekly active at day 30 | 110 |
| Stayed solo | Lapsed | 210 |
The same data as CSV
source,target,value
Accounts created,Verified email,820
Accounts created,Never verified,180
Verified email,Created first project,560
Verified email,Did not start a project,260
Created first project,Invited a teammate,240
Created first project,Stayed solo,320
Invited a teammate,Weekly active at day 30,180
Invited a teammate,Lapsed,60
Stayed solo,Weekly active at day 30,110
Stayed solo,Lapsed,210How it works
List the first steps
Choose the three or four actions that define a successfully onboarded account.
Choose the split
Pick the one behaviour you want to compare, such as inviting a teammate, and place it after the step it follows.
Count a single cohort
Use accounts created in one period and look at day-30 activity for all of them.
Enter and check
Type the counts and make sure every step sends out what it received.
Frequently asked questions
Why end both branches in the same outcome nodes?
So the widths are comparable. If the branches ended in different nodes you could not see at a glance which one retains better.
Does the diagram prove the invitation matters?
No. It shows how accounts distribute across paths. Testing a cause needs an experiment or a controlled comparison.
Edit this onboarding funnel
Open the example as your own chart and enter your onboarding counts.