Jump to content

Research:Account Creation Referrals

From Meta, a Wikimedia project coordination wiki
Dario Taraborelli
Brandon Harris

This page documents a planned research project.
Information may be incomplete and change before the project starts.

Key Personnel[edit]

  • Dario Taraborelli, Wikimedia Foundation
  • Howie Fung, Wikimedia Foundation
  • Brandon Harris, Wikimedia Foundation


E3 Experiments is tracking referrals to the "create account" page, though not the protection level of referring wiki pages.

Project Summary[edit]

With some notable exceptions (such as creating new articles or editing [semi-]protected articles), there is no need for a visitor to register an account in order to edit a Wikipedia article. Registering an account provides editors with the ability to track their individual contributions, to hide their IP address as well as other privileges that are tied to registered accounts.

The goal of this project is to better understand why people register an account on a Wikimedia project and to analyze the different levels of activity and survival rates of accounts that were registered via specific campaigns (e.g. a global education initiative or an outreach event), funnels (e.g. a call-to-action triggered by article feedback) or type of internal referral (e.g. a high-traffic article, a semi-protected page, an article undergoing an edit war). This project is complementary to the Account creation user experience and earlier Account Creation Improvement Projects in that it focuses on the factors that determine the intention to create a new account rather than the effectiveness of different account creation processes at successfully completing new account registrations.


Wikimedia Policies, Ethics, and Human Subjects Protection[edit]

Data collection will be performed under the terms of the Wikimedia Privacy Policy and after full clearing of the proposed method by the Foundation's legal team.

Benefits for the Wikimedia community[edit]

We have a poor/anecdotical understanding of the reasons why new editors join and register an account. Data collected from MoodBar suggests that high-traffic or semi-protected articles are among the top sources from which new editors post their mood, which in turn suggests that these may be the most popular articles new editors intend to edit and the most popular internal referrals of new accounts. This project will allow us to quantify how many users register an account originating from a semi-protected article just to find out that they cannot edit because they are not yet autoconfirmed. It will also allow us to understand what happens to all those users who sign up but never complete their first edit: is this because they get lost in the process or because they cannot edit due to technical restrictions such as semi-protection. Finally, this project will allow us to study the survival of different cohorts of new editors and the effectiveness of different initiatives/campaigns/funnels at attracting high-retention editors.

Time Line[edit]


This is an internal project of the Wikimedia Foundation.

See also[edit]