Product & Strategy 8 projects

Other collectionsBuilt 3Claude in Daily Life 2Strategy Decks 2PRD Library 5

Product analytics case study

Fumble

Comparing retention across Standard and Pro plans.

I analyzed July to December 2018 monthly cohorts to compare retention behavior across Fumble’s two subscription plans and surface plan-level patterns relevant to an investor discussion.

  • Product Analytics
  • Cohort Analysis
  • Retention
  • Python
monthly cohorts
6
subscription plans
2
analysis window
6-month
users in cohorts
2,013
8 sections · 2 original artifacts

Fumble is a dating app with two subscription plans, Standard and Pro. The exercise asked for the July to December 2018 monthly cohorts to be analyzed over their first six months for retention and average revenue, with the results prepared for an investor conversation.

What this isThis is a product exercise on a supplied dataset. It reports what the data shows and does not claim why, and no product change was shipped.

The ideaA cohort analysis of Fumble's Jul to Dec 2018 subscribers: retention by month for the Standard and Pro plans, compared cohort by cohort, with every number recomputed from the workbook.

01

Six cohorts per plan

4,711 subscribers were grouped by plan and start month into six monthly cohorts, each followed for months 0 to 5.

The method

02

Retention, recomputed

A subscriber counts as retained at month m when their active months exceed m. Every chart is recomputed from the workbook, never typed.

The method

03

Standard fades steadily

Pooled across its six cohorts, Standard retention falls from 100% at month 0 to 32.4% at month 5.

Standard retention

04

Pro holds on longer

Pooled across its six cohorts, Pro retention falls from 100% at month 0 to 46.5% at month 5.

Pro retention

05

Pro leads in every cohort

Pro retains more than Standard at month 5 in 6 of 6 cohorts, compared cohort by cohort.

The comparison

06

The data is on the page

The supplied workbook sits behind every chart, with a preview of its rows and a download.

The source data

Standard Plan: 1,391 users across six cohorts

Standard Plan: % of each cohort still active
Standard Plan: % of each cohort still active. Values are % of the cohort. Rows are cohorts, columns are Retention month.
Retention month
CohortUsers012345
Jul ’18151100.090.180.868.260.957.6
Aug ’18160100.089.473.863.857.554.4
Sep ’18234100.082.980.365.860.346.6
Oct ’18279100.090.775.364.941.625.1
Nov ’18260100.075.872.351.922.321.5
Dec ’18307100.095.464.516.015.313.4

Values are % of the cohort. Darker tint means a higher value.

Standard Plan: retention by month
0%20%40%60%80%100%012345Months since joining87.473.652.039.332.40%20%40%60%80%100%012345Months since joining8774523932
Standard Plan: retention by month: Standard Plan retention by month for the six Jul to Dec 2018 cohorts. Pooled across cohorts, retention falls from 100% at month 0 to 32.4% at month 5.
Retention monthJul ’18 cohortAug ’18 cohortSep ’18 cohortOct ’18 cohortNov ’18 cohortDec ’18 cohortStandard Plan, all six cohorts pooled
0100.0%100.0%100.0%100.0%100.0%100.0%100.0%
190.1%89.4%82.9%90.7%75.8%95.4%87.4%
280.8%73.8%80.3%75.3%72.3%64.5%73.6%
368.2%63.8%65.8%64.9%51.9%16.0%52.0%
460.9%57.5%60.3%41.6%22.3%15.3%39.3%
557.6%54.4%46.6%25.1%21.5%13.4%32.4%

The ink line pools all six cohorts. Each tinted line is one cohort; point at the chart to read exact values.

Pro Plan: 622 users across six cohorts

Pro Plan: % of each cohort still active
Pro Plan: % of each cohort still active. Values are % of the cohort. Rows are cohorts, columns are Retention month.
Retention month
CohortUsers012345
Jul ’1885100.096.592.988.272.971.8
Aug ’1885100.098.892.985.976.572.9
Sep ’1895100.092.686.376.865.353.7
Oct ’18101100.098.085.171.360.444.6
Nov ’18131100.078.672.554.229.029.0
Dec ’18125100.099.272.029.628.025.6

Values are % of the cohort. Darker tint means a higher value.

Pro Plan: retention by month
0%20%40%60%80%100%012345Months since joining93.282.264.551.946.50%20%40%60%80%100%012345Months since joining9382645246
Pro Plan: retention by month: Pro Plan retention by month for the six Jul to Dec 2018 cohorts. Pooled across cohorts, retention falls from 100% at month 0 to 46.5% at month 5.
Retention monthJul ’18 cohortAug ’18 cohortSep ’18 cohortOct ’18 cohortNov ’18 cohortDec ’18 cohortPro Plan, all six cohorts pooled
0100.0%100.0%100.0%100.0%100.0%100.0%100.0%
196.5%98.8%92.6%98.0%78.6%99.2%93.2%
292.9%92.9%86.3%85.1%72.5%72.0%82.2%
388.2%85.9%76.8%71.3%54.2%29.6%64.5%
472.9%76.5%65.3%60.4%29.0%28.0%51.9%
571.8%72.9%53.7%44.6%29.0%25.6%46.5%

The ink line pools all six cohorts. Each tinted line is one cohort; point at the chart to read exact values.

From subscriber rows to cohort tables

  1. Raw subscription data

    4,711 subscribers with plan, start month and active months.

  2. Cohort assignment

    Six monthly cohorts per plan, grouped by start month, Jul to Dec 2018.

  3. Active-month calculation

    Retained at month m when active months exceed m. Month 0 is the cohort size.

  4. Cohort retention table

    Counts and percentages for months 0 to 5, built in Python.

  5. Plan comparison

    Standard and Pro compared cohort by cohort at month 5.

Data snapshotThe charts on this page are computed from the current workbook. The original one-page analysis was built from an earlier data snapshot and differs slightly for the July cohort (Standard 145 vs 151 users, Pro 84 vs 85), so its numbers are kept out of these charts rather than mixed in.

RevenueThe brief also asked for average revenue. The workbook holds a lifetime revenue total per subscriber with no stated first-six-month method, so no revenue figure is shown rather than assuming one.

Month-5 retention by cohort
0%20%40%60%80%100%Jul ’18Aug ’18Sep ’18Oct ’18Nov ’18Dec ’18Monthly cohort71.872.953.744.629.025.657.654.446.625.121.513.40%20%40%60%80%100%Jul ’18Aug ’18Sep ’18Oct ’18Nov ’18Dec ’18Monthly cohort727354452926585447252213
Month-5 retention by cohort: Month-5 retention by cohort. Pro Plan: Jul 71.8%, Aug 72.9%, Sep 53.7%, Oct 44.6%, Nov 29.0%, Dec 25.6%. Standard Plan: Jul 57.6%, Aug 54.4%, Sep 46.6%, Oct 25.1%, Nov 21.5%, Dec 13.4%.
CategoryPro PlanStandard Plan
Jul ’1871.8%57.6%
Aug ’1872.9%54.4%
Sep ’1853.7%46.6%
Oct ’1844.6%25.1%
Nov ’1829.0%21.5%
Dec ’1825.6%13.4%

Share of each cohort still active at retention month 5, from the current workbook.

The gap at month 5, in percentage points

Month-5 retention by cohort: Pro Plan, Standard Plan and the gap between them in percentage points.
CohortProStandardGap
Jul ’1871.8%57.6%+14.1 pts
Aug ’1872.9%54.4%+18.6 pts
Sep ’1853.7%46.6%+7.1 pts
Oct ’1844.6%25.1%+19.5 pts
Nov ’1829.0%21.5%+7.5 pts
Dec ’1825.6%13.4%+12.2 pts

Standard-plan cohorts contain more users in absolute terms.

These are descriptive observations. The data does not explain why Pro users stay longer, and later cohorts have had less time to accumulate months, so a cohort-by-cohort read is the fairest comparison.

Tumble: Report a user

From the same course file. It is not part of the cohort analysis.

Open in Figma (opens in a new tab)

1 / 4

Step 01

Home: report entry

A profile card in the Tumble dating app carries a Report button next to the person's name.

  • Entry pointReport beside the profile name
  • Also on screenSuper Like and Mode
  • Bottom barProfile, Bees, Likes, Chat

PMF: Week 5 Project Data Set

XLSX · 4,711 rows · 7 columns

Subscriber rows with plan, start month and active months.

Download data (opens in a new tab)
IDJoin DateEnd DatePlan TypeRevenueStart MonthActive Months
899502018-01-252019-03-02Standard Plan$160.00Jan-201814
899512018-01-262018-03-27Standard Plan$40.00Jan-20182
899522018-01-262021-01-27Pro Plan$1,234.29Jan-201837
899532018-01-262021-01-27Pro Plan$1,234.29Jan-201837
899542018-01-272021-01-27Pro Plan$1,234.29Jan-201837
899552018-01-272021-07-11Standard Plan$461.71Jan-201843
899582018-01-292018-10-26Standard Plan$180.00Jan-20189
899842018-02-032018-04-19Standard Plan$40.00Feb-20183

First 8 of 4,711 rows. The full workbook is not embedded in the page.

Original artifacts2

The raw data behind every chart, and the related prototype file.

Download the original documents (zip)
Next in Product & Strategy · 3 of 8ShareChat KreatorHow AI-assisted regional-language content creation could help ShareChat users create and share better posts.Next project

Sections 8

Original artifacts2