Takehome exercise provided by Asana for Data Science internship. Suggested time for completion: ~3 hours

Provided Information along with data: We define an "adopted user" as a user who has logged into the product on three separate days in at least one seven-day period. Because we believe that adopted users are more likely to be successful at using Asana in the long term than those that are not adopted, we want to know what things are likely indicators of future adoption. Objective: With this in mind, we'd like you to identify which factors predict user adoption.

The data has the following two files:

A user file ("takehome_users") with data on 12,000 users who signed up for the product in the last two years. This table includes:

• name: the user's name
• object_id: the user's id
• email: email address
• email_domain: domain of email address, e.g. gmail.com
• creation_source: how they signed up for the product. This takes on one of 5 values:
• PERSONAL_PROJECTS: invited to join another user's personal workspace
• GUEST_INVITE: invited to an organization as a guest (limited permissions)
• ORG_INVITE: invited to an organization (as a full member)
• SIGNUP: signed up via asana.com
• SIGNUP_GOOGLE_AUTH: signed up using Google
• Authentication (using a Google email account for their login id)
• creation_time: when they created their account
• last_session_creation_time: unix timestamp of last login
• opted_in_to_mailing_list: whether they have opted into receiving marketing emails
• enabled_for_marketing_drip: whether they are on the regular marketing email drip
• org_id: the organization (group of users) they belong to
• invited_by_user_id: which user invited them to join (if applicable).

(second data file)

• A usage summary file ("takehome_user_engagement") that has a row for each day that a user logged into the product.

Please send us a brief report with your findings, along with any summary tables or graphs that you think will help us better understand them. We will not consider more than 1 page of text and 1 page of supplementary visuals. Please also include all of the code you used to arrive at your results. We will be grading the report and your code.

Project walkthrough

• Part 1 understanding the data:

  • Raw Data exploration("takehome_users"): Of the data provided, I am thinking that the most useful will be creation_source (to see which source leads to user adoption; marketing drip and mailing list usage which could demonstrate which users may be interested to learn more about Asana and how to integrate it into their workflow; creation time which may identify starting point of user's adoption of Asana; and last_session_creation_time which is filled with 10 digit entries and has missing values for various users. I will assume this is correctly provided data, however I am unsure how I'll be able to apply this information in a useful manner.

  • Raw Data exploration("takehome_user_engagement"): Raw data seems to indicate the amount of times a user (identified by user ID) engages with Asana with respected dates and time. The time_stamp column includes both date and time which may cause issues when parsing through. This data may have to be separated for easier analysis of which users may be "adopted users" based on their engagement in a 7-day period.

• Part 2 Data cleaning

• Part 3 Feature Space Selection

• Part 4 Data Modeling

• Part 5 Optimization and Deployment