Country or region and language

    Choose your country or region and your language.

    Suggested for you

      All countries and regions

        Choose your character

        It shows in the top bar and on your profile. Change it any time.

        Projects/Pennant

        Who cancels in the first 90 days?

        Pennant loses too many subscribers in their first three months. Build a churn table without peeking at the future, find who leaves, fit a first model and plan what the retention team should try.

        ProjectPennant · fictional company2 h 50 min4 stagesData and analytics

        One brief, 3 toolsYour tables pass from SQL to pandas to a scikit-learn notebook. You finish with a retention plan.

        Pennant

        Fitness app

        About the company

        A fitness app with guided workouts, sold as a monthly or annual subscription in the UK.

        Pennant gives people guided workouts and training plans they can follow at home or in the gym. Subscribers pay by the month or for a year up front, and they sign up through the app stores, a friend's referral or a paid social ad. They live all over the UK, including London, the Midlands, Scotland and Wales. About one in four new monthly subscribers cancels within three months. You work in the data team with two groups: retention, which tries to keep new subscribers training, and product, which runs tests in the checkout.

        More at Pennant

        Founded
        2019
        Head office
        Manchester, UK
        Plans
        Monthly or annual
        Employees
        About 70

        Who writes the briefs

        • Hannah OkoroHead of Retention
        • Tom AshbyProduct Lead

        A fictional company, invented for practice. Meet the other companies

        The brief

        From
        Hannah OkoroHead of Retention
        To
        Data team
        Subject
        Early cancellations

        About one in four new monthly subscribers cancels within three months, and the retention team has budget for one intervention next quarter: an onboarding call, a discount, or a training plan in the app. I want to know who is most likely to leave early and whether we can spot them in their first month, while there is still time to help.

        You have every subscriber who joined between April and October, their weekly workouts and the date anyone cancelled. The data was cut on October 31, so people who joined recently have not had 90 days yet.

        Done looks like: one row per subscriber with what we knew in their first 28 days and whether they left within 90, churn rates by plan and early activity, a first model tested on later signups, and a plan for the retention team.

        • Data: 260 subscribers, their weekly workouts and cancellations, April to October 2026.
        • You deliver a churn table, churn rates by activity, a tested model and a retention plan.
        • Checks cover subscribers too new to measure, activity after day 28, and a date-based test split.

        The project path

        1. 1
          SQL45 min

          Build one row per subscriber

          One row per subscriber with their first-month activity and whether they cancelled within 90 days, leaving out anyone too new to tell.

          Reads
          subscribers · workouts
          Produces
          churn_features

          pandas reads the churn table.

          Open stageOne Course in SQL or All Access
        2. 2
          pandas35 min

          Churn by early activity

          Band subscribers by how much they trained in their first four weeks and compare churn by plan.

          Reads
          churn_features
          Produces
          churn_by_band

          Your plan cites the churn rates.

          Open stageOne Course in pandas or All Access
        3. 3
          Machine learning45 min

          Fit a first churn model

          Fit a logistic regression on earlier signups and score it on later ones.

          Reads
          churn_features
          Produces

          Your plan says how far the model can be trusted.

          Open stageOne Course in Machine learning or All Access
        4. 4
          Plan25 min

          Write the retention plan

          Say who leaves, how early you can tell, which intervention to try first and how to measure it.

          Reads
          churn_by_band
          Produces
          retention_plan.md

          Optional: present it in Interview Hub.

          Open stageAll Access

        Each stage includes reference inputs. Starting in the middle or using them doesn't complete the earlier stages.

        Project access

        Free

        $0

        Read all briefs and paths. Tool stages and written work are locked.

        Keep browsing

        One Course

        $17.99/month

        Practice the stages in your chosen tool. Reference tables let you start at that stage.

        All Access

        $29.99/month

        Every stage in every format, plus optional spoken readouts using dumbbells.

        Your Work Gym account

        Save your progress and pick up on another device. Your browser's progress comes with you.

        or

        You can keep practicing without an account.

        Help and feedback

        Ask a question, tell us what is not working, or share an idea. We reply by email.

        What is it about?

        We use it only to reply to you.