Cohort analysis patterns in Tableau - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ dax_lod_result

intermediate

Calculate Monthly Cohort Size Using LOD Expression

You have a dataset with customer sign-up dates and purchase dates. Which Tableau LOD expression correctly calculates the number of customers who signed up in each month (cohort size)?

A { INCLUDE DATETRUNC('month', [Purchase Date]) : COUNTD([Customer ID]) }

B { EXCLUDE [Customer ID] : COUNTD([SignUp Date]) }

C { FIXED [Customer ID] : COUNTD(DATETRUNC('month', [SignUp Date])) }

D { FIXED DATETRUNC('month', [SignUp Date]) : COUNTD([Customer ID]) }

💡 Hint

Think about fixing the cohort by sign-up month and counting unique customers.

The correct LOD expression fixes the calculation at the month level of the sign-up date and counts distinct customers. Option D does this correctly.

Option D fixes by purchase month, which is not the cohort definition.

Option D excludes customer ID incorrectly and counts sign-up dates, which is wrong.

Option D fixes by customer ID, which is not the cohort grouping.

❓ visualization

intermediate

Best Visualization for Monthly Cohort Retention

You want to show how many customers from each monthly cohort make repeat purchases over the next 6 months. Which visualization type best communicates this pattern in Tableau?

A Line chart with purchase dates on X-axis and total sales on Y-axis

B Pie chart showing total customers per cohort

C Heatmap with cohorts on rows and months since sign-up on columns, colored by retention count

D Bar chart showing total purchases per month

💡 Hint

Think about showing retention over time for each cohort distinctly.

A heatmap with cohorts as rows and months since sign-up as columns colored by retention count clearly shows retention patterns over time.

Pie charts and bar charts do not show retention over time per cohort.

Line charts with purchase dates show overall sales trends, not cohort retention.

🧠 Conceptual

advanced

Understanding Cohort Analysis Purpose

Why is cohort analysis important in business intelligence when analyzing customer behavior?

A It groups customers by shared characteristics to track behavior over time, revealing retention and lifecycle patterns.

B It aggregates all customers into one group to calculate total sales only.

C It predicts future sales using only last month’s data.

D It removes outliers from the dataset to improve accuracy.

💡 Hint

Think about how grouping by time of first action helps understand behavior changes.

Cohort analysis groups customers by shared starting points (like sign-up month) to observe how their behavior changes over time, which helps understand retention and lifecycle.

Options B, C, and D do not describe cohort analysis correctly.

❓ data_modeling

advanced

Designing Data Model for Cohort Analysis

Which data model design best supports efficient cohort analysis in Tableau?

A A fact table with customer transactions linked to a customer dimension table containing sign-up date

B A single flat table with all data duplicated for each transaction without keys

C Separate tables for sales and marketing with no relationship

D Only aggregated monthly sales data without customer details

💡 Hint

Think about how to connect transactions to customer cohorts efficiently.

Option A provides a normalized model linking transactions to customer info including sign-up date, enabling cohort grouping.

Option A duplicates data causing inefficiency.

Option A lacks relationships needed for analysis.

Option A lacks customer-level data needed for cohorts.

🔧 Formula Fix

expert

Identify the Error in Cohort Retention Calculation

You wrote this Tableau calculated field to compute retention rate for each cohort month:

SUM([Repeat Purchases]) / COUNTD([Customer ID])

But the retention rates are incorrect and inconsistent. What is the most likely cause?

Tableau

SUM([Repeat Purchases]) / COUNTD([Customer ID])

A COUNTD([Customer ID]) counts repeat purchases instead of unique customers.

B The denominator counts all customers, not just those in the cohort month, causing incorrect retention rates.

C The formula should use AVG instead of SUM for repeat purchases.

D SUM aggregation cannot be used on [Repeat Purchases] because it is a string field.

💡 Hint

Check if the denominator matches the cohort group correctly.

The denominator must count only customers in the specific cohort month to calculate correct retention. Counting all customers inflates the denominator and skews rates.

Option B is incorrect because repeat purchases is numeric.

Option B is wrong because SUM is appropriate for counting repeat purchases.

Option B is incorrect because COUNTD counts unique customers, not purchases.