Background

We will run an experiment targeting logged out users on mobile web, showing the treatment group a bookmark "save" button at the top of an article instead of the "watch star" in control group. When clicked, a modal will inform users that if they create an account or log in, they will be able to save articles to a reading list.

We need a measurement plan to compare the account creation rate between the control and treatment groups to see whether the treatment has a higher rate. (Goal: 3% relative improvement compared to control)

Some interesting secondary metrics we should track:

• The click rate of "bookmark" button in treatment
• The click through rate on the "create account" button in treatment (as it can be different from account creation rate)
• The click through rate of "log in" to existing account on treatment
• The click rate of "watch star" button in control
• The click rate of "create account" in control
• The click rate of "log in" in control

Targeting considerations
Due to concurrent experiments being run by the Reader Growth (on "Share" button replacing the "download" button for logged out mobile web users, wherein they're also using a pulsating dot to attract user attention), and by Editor Growth (an experiment with streamlining the account creation flow, which might interfere with our results), we should avoid targeting users in the following wikis for this experiment:

• enwiki (Editor Growth experiment, WE 1.8.3)
• ar, id, vi, zh, fr (Reader Growth experiment, WE 3.10.5) -- might change, to be updated.

Design requirements

See the user flows here.

---

Requirements

Acceptance criteria

• We have a measurement plan (Kirsten has put all the Account Creation measurement plans into the same document, on different document tabs, in case we want to also create our plan there)
• We have a measurement spec

BDD

• For QA engineer to fill out.

Test Steps

• For QA engineer to fill out.

Communication criteria

Add if this needs an announcement or discussion.

Rollback plan

Describe the rollback plan in production for this task if something goes wrong.

This task was created by Version 1.0.0 of the Reader Experience team task template using phabulous.