Delay workspace setup until after first value

Users who reach workspace setup before their first project activate less often than users who bypass it. In the example journey the front page draws, 10,000 visits become 4,400 signups, 3,900 of those reach setup, 1,900 create a first project and 1,400 pay. The numbers are the blueprint's example and not a customer's; the shape is the point: the largest loss sits at the one step that asks for work before it has shown any value.

Instructions and configuration a user has to get through before using a product cost attention and effort, reduce usability, and should be avoided where possible [1]. A required setup step before the first project is exactly that cost, placed at the moment the product has not yet proved itself. The loop that found this leak, and the way the example funnel is drawn, is on the how it works page.

New workspaces created on the self-serve path. Invited teammates are excluded, because they join a workspace that is already set up, and enterprise trials are excluded, because their setup is done with a person.

Making setup optional and moving it after the first project raises paid activation, because it removes a blocking step that sits before the user has seen the product do anything, and a user who has made something has a reason to configure the place it lives.

Control

The workspace setup screen is required to continue: a user cannot reach the first project until it is completed, including the invite your team step.

Variant

The setup screen is optional and can be skipped in one click. After the first project is created, a one-line prompt asks the user to finish setup, dismissible, and setup stays reachable from the workspace menu.

Name

Paid activation

Definition

The share of eligible new workspaces that start a paid plan

Window

14 days from signup

Secondary: First project created

The share of eligible new workspaces with a first project, 24 hours from signup.

Secondary: Setup completed

The share of eligible new workspaces that complete setup, 14 days from signup.

Support volume

Must not rise. Support conversations per hundred eligible new workspaces in the window

Setup completion within 30 days

Must not drop. The share of eligible new workspaces that complete setup within 30 days, against its baseline share

Cancellation rate

Must not rise. The share of newly paid workspaces that cancel within their first cycle

This is an example. No expected impact is stated because there is no evidence to state one from.

Allocation

20 percent of eligible new workspaces to the variant, 80 percent to control

Eligibility

New self-serve workspaces from the moment the flag is on; a workspace keeps its arm for its lifetime

Ramp

5 percent for the first days to catch a broken flow, then 20 percent to the planned sample

Minimum sample

Size the test before it starts: the minimum sample follows from the baseline activation rate and the smallest lift worth acting on, and a test that cannot reach it at this traffic in a reasonable time should not run.

Size it with

A/B test sample size calculator

The required gate becomes an optional prompt behind the customer's existing feature flag. The setup component, its copy and the prompt are the customer's own; the variant reorders when they appear and adds a skip. Exposure is recorded at the moment the setup screen would have been shown, so the two arms are compared on the same population.

Ship when the primary metric's interval clears zero at the planned sample with every guardrail holding. Revert on any guardrail breach, at any point. Otherwise continue to the planned sample without peeking, and call the result inconclusive if it does not clear zero there.