Interview Prep
Career
GenAI
Prompt Engg
ChatGPT
LLM
Langchain
RAG
AI Agents
Machine Learning
Deep Learning
GenAI Tools
LLMOps
Python
NLP
SQL
AIML Projects

Reading list

Machine Learning Basics for a Newbie
6 Steps of Machine learning Lifecycle Introduction to Predictive Modeling
Introduction to Exploratory Data Analysis & Data Insights Descriptive Statistics Inferential Statistics How to Understand Population Distributions?
Reading Data Files into Python Different Variable Datatypes
Probability for Data Science Basic Concepts of Probability Axioms of Probability Conditional Probability
Central Tendencies for Continuous Variables Spread of Data KDE plots for Continuous Variable Overview of Distribution for Continuous variables Normal Distribution Skewed Distribution Skeweness and Kurtosis Distribution for Continuous Variable
Central Tendencies for Categorical Variables Understanding Discrete Distributions Performing EDA on Categorical Variables
Dealing with Missing Values Understanding Outliers Identifying Outliers in Data Outlier Detection in Python Outliers Detection Using IQR, Z-score, LOF and DBSCAN
Sample and Population Central Limit Theorem Confidence Interval and Margin of Error
Bivariate Analysis Introduction
Covariance Pearson Correlation Spearman's Correlation & Kendall's Tau Correlation versus Causation Tabular and Graphical methods for Bivariate Analysis Performing Bivariate Analysis on Continuous-Continuous Variables
Tabular and Graphical methods for Continuous-Categorical Variables Introduction to Hypothesis Testing P-value Two sample Z-test T-test T-test vs Z-test Performing Bivariate Analysis on Continuous-Catagorical variables
Chi-Squares Test Bivariate Analysis on Categorical Categorical Variables
Multivariate Analysis A Comprehensive Guide to Data Exploration The Data Science behind IPL
Supervised Learning vs Unsupervised Learning Reinforcement Learning Generative and Descriminative Models Parametric and Non Parametric model
Machine Learning Pipeline Preparing Dataset Build a Benchmark Model: Regression Build a Benchmark Model: Classification
Evaluation Metrics for Machine Learning Everyone should know Confusion Matrix Accuracy Precision and Recall AUC-ROC Log Loss R2 and Adjusted R2
Dealing with Missing Values Replacing Missing Values Imputing Missing Values in Data Working with Categorical Variables Working with Outliers Preprocessing Data for Model Building
Understanding Cost Function Understanding Gradient Descent Math Behind Gradient Descent Assumptions of Linear Regression Implement Linear Regression from Scratch Train Linear Regression in Python Implementing Linear Regression in R Diagnosing Residual Plots in Linear Regression Models Generalized Linear Models Introduction to Logistic Regression Odds Ratio Implementing Logistic Regression from Scratch Introduction to Scikit-learn in Python Train Logistic Regression in python Multiclass using Logistic Regression How to use Multinomial and Ordinal Logistic Regression in R ? Challenges with Linear Regression Introduction to Regularisation Implementing Regularisation Ridge Regression Lasso Regression
Introduction to K Nearest Neighbours Determining the Right Value of K in KNN Implement KNN from Scratch Implement KNN in Python
Bias Variance Tradeoff Introduction to Overfitting and Underfitting Visualizing Overfitting and Underfitting Selecting the Right Model What is Validation? Hold-Out Validation Understanding K Fold Cross Validation
Introduction to Feature Selection Feature Selection Algorithms Missing Value Ratio Low Variance Filter High Correlation Filter Backward Feature Elimination Forward Feature Selection Implement Feature Selection in Python Implement Feature Selection in R
Introduction to Decision Tree Purity in Decision Tree Terminologies Related to Decision Tree How to Select Best Split Point in Decision Tree? Chi-Squares Information Gain Reduction in Variance Optimizing Performance of Decision Tree Train Decision Tree using Scikit Learn Pruning of Decision Trees
Introduction to Feature Engineering Feature Transformation Feature Scaling Feature Engineering Frequency Encoding Automated Feature Engineering: Feature Tools
Introduction to Naive Bayes Conditional Probability and Bayes Theorem Introduction to Bayesian Adjustment Rating: The Incredible Concept Behind Online Ratings! Working of Naive Bayes Math behind Naive Bayes Types of Naive Bayes Implementation of Naive Bayes
Understanding how to solve Multiclass and Multilabled Classification Problem Evaluation Metrics: Multi Class Classification
Introduction to Ensemble Techniques Basic Ensemble Techniques Implementing Basic Ensemble Techniques Finding Optimal Weights of Ensemble Learner using Neural Network Why Ensemble Models Work well?
Introduction to Stacking Implementing Stacking Variants of Stacking Implementing Variants of Stacking Introduction to Blending Bootstrap Sampling Introduction to Random Sampling Hyper-parameters of Random Forest Implementing Random Forest Out-of-Bag (OOB) Score in the Random Forest IPL Team Win Prediction Project Using Machine Learning Introduction to Boosting Gradient Boosting Algorithm Math behind GBM Implementing GBM in python Regularized Greedy Forests Extreme Gradient Boosting Implementing XGBM in python Tuning Hyperparameters of XGBoost in Python Implement XGBM in R/H2O Adaptive Boosting Implementing Adaptive Boosing LightGBM Implementing LightGBM in Python Catboost Implementing Catboost in Python
Different Hyperparameter Tuning methods Implementing Different Hyperparameter Tuning methods GridsearchCV RandomizedsearchCV Bayesian Optimization for Hyperparameter Tuning Hyperopt
Understanding SVM Algorithm SVM Kernels In-depth Intuition and Practical Implementation SVM Kernel Tricks Kernels and Hyperparameters in SVM Implementing SVM from Scratch in Python and R
Introduction to Principal Component Analysis Steps to Perform Principal Compound Analysis Computation of Covariance Matrix Finding Eigenvectors and Eigenvalues Implementing PCA in python Visualizing PCA A Brief Introduction to Linear Discriminant Analysis Introduction to Factor Analysis
Introduction to Clustering Applications of Clustering Evaluation Metrics for Clustering Understanding K-Means Implementation of K-Means in Python Implementation of K-Means in R Choosing Right Value for K Profiling Market Segments using K-Means Clustering Hierarchical Clustering Implementation of Hierarchial Clustering DBSCAN Defining Similarity between clusters Build Better and Accurate Clusters with Gaussian Mixture Models
Understand Basics of Recommendation Engine with Case Study
8 Ways to Improve Accuracy of Machine Learning Models
Introduction to Dask Working with CuML
Introduction to Machine Learning Interpretability Framework and Interpretable Models model Agnostic Methods for Interpretability Implementing Interpretable Model Understanding SHAP Out-of-Core ML Introduction to Interpretable Machine Learning Models Model Agnostic Methods for Interpretability Game Theory & Shapley Values
Introduction to Machine Learning Interpretability Framework and Interpretable Models model Agnostic Methods for Interpretability Implementing Interpretable Model Understanding SHAP Out-of-Core ML Introduction to Interpretable Machine Learning Models Model Agnostic Methods for Interpretability Game Theory & Shapley Values
Introduction to AutoML Implementation of MLBox Introduction to PyCaret TPOT Auto-Sklearn EvalML
Pickle and Joblib Introduction to Model Deployment
Deploying Machine Learning Model using Streamlit Deploying ML Models in Docker Deploy Using Streamlit Deploy on Heroku Deploy Using Netlify Introduction to Amazon Sagemaker Setting up Amazon SageMaker Using SageMaker Endpoint to Generate Inference Deploy on Microsoft Azure Cloud Introduction to Flask for Model Deploying ML model using Flask
Model Deployment in Android Model Deployment in Iphone

Business Case Study Assignments For Entry Level Data Analysts

Chrisx10
Chrisx10 Last Updated : 12 Oct, 2024
8 min read

Introduction

Why do companies conduct business case studies along with interviews? Why not just be done with interviews and save time and effort? Companies want to hire data analysts who can apply theoretical principles to solve practical problems, find solutions, and be deductive. Not everyone is deductive, and most people are inductive; they learn from experiences, anecdotes, observations, and patterns. Case studies can gauge the inductive ability of candidates, help identify problem solvers, and make healthy hiring decisions.

In the real world, unlike bootcamps and Coursera mini projects, data is messy, scarce, sporadic, sparse, incoherent, and very rarely clean as expected. When data is clean, it calls for a deeper understanding of where it was sourced from, how it can be leveraged, etc. So companies would ideally want to understand what a data analyst would do in this scenario, where nothing seems right and everything is ambiguous.

Most data analysts solve open-ended problems with multiple open-ended solutions. There is no one right solution; if someone says there is one, it could be out of ignorance and incompetence. Data analysts are expected to 1. Give life to a new analysis(0 to 1), 2. Work on something that’s already existing and enhance it, 3. Repeat the same task daily, weekly, monthly, etc., 4. Solve new problem statements daily, weekly, and monthly. Broadly this converts into four combinations:

data analyst

Learning Objectives:

  1. Describe the case study. E.g., Business sales, Loans, etc.
  2. Identify the key issues and problems.
  3. Analyze the case study using theoretical principles of statistics and data science.
  4. Recommend the right course of action for each issue.

This article was published as a part of the Data Science Blogathon.

Table Of Contents

  1. Advantages of Using Case Studies
  2. Guidelines to Solve Case Studies
  3. Sales Leads and Conversions
  4. Problem Statement- Improving the Efficiency of Sales Conversion
  5. Our Recommendation
  6. Useful Resources and References
  7. Conclusion

Advantages of Using Case Studies for Data Analysts

  • Problem-solving (PS) – PS plays an important role for DS (data scientists)/DA (data analyst)/BA (business analyst). The magnitude of efforts used on PS can vary across organizations and projects, with some teams working mainly on PS about 80% of the time, while others work 20% of the time.
  • Analytical Tools, Quantitative And/or Qualitative, Depending on the Case – Various techniques and approaches are available, and deciding which method/framework fits the bill also depends on the data analyst.
  • Decision-making in Complex Situations – With multiple stakeholders churning out multiple solutions, the final say rests with the data analyst. Turning complex situations into easy-to-understand business solutions is one of the key abilities of a data analyst.
  • Coping with Ambiguities – Most problems that need to be solved are not clear beforehand; sometimes, a lot of effort goes into framing the problem statement before actually trying to solve it using ML models or analytics.

Business case studies are excellent opportunities for candidates to showcase these abilities and stand out from the crowd. Here problem-solving analysts differentiate themselves from task-driven data analysts.

Guidelines to Solve Case Studies for Data Analysts

  1. What is the issue?
    1. Identifying the reason for the analysis provides a lot of clarity on how to do the analysis, knowing the issue at hand.
    2. Issues can be how to reduce losses(customers loss, growth loss, subscriber loss, monetary loss, etc.), or optimize a metric, for example, sales conversion. 
    3.  How to grow sales conversion by X percent in the next 6 months can also be considered an issue. 
    4. Any objective that has a positive business impact can be considered. 
  2. What is the goal of the analysis?
    1. Once the issue is known, the goal is to use data science techniques to solve it.
    2. There can be multiple goals as well; for example: in the North, grow sales by X%, and in the South, grow sales by Y% percentage. The same technique can be used to reach these 2 goals.
    3. Visualize the end goal in a tabular form or as a flowchart. This will provide a lot of clarity while coding and solving it.
  3. What is the context of the problem?
    1. Why solve this problem?
    2. What’s the overall impact(for example – 35Cr profit lift for the month of Jan 23)?
    3. Which business will be impacted?
    4. Who is the stakeholder?
    5. What is the budget/cost/investment involved?
  4. What would you recommend — and why?
    1. Take a stand, and prove your point, backed with data.
    2. Recommendations are practical ideas based on insights.
    3. Provide multiple recommendations and provide pros and cons for each.

Finally, even though the final recommendation is important, the approach and reasoning to get there are also equally important.

Sales Leads and Conversions

data analyst
Source: https://stevepatrizi.com/

The sales funnel a prospective customer’s path to convert into a customer. At the top, there are a lot of users (to be customers), and as the funnel moves down, the size of the funnel reduces. This is where the sales team comes into the picture. They convert these potential customers into customers and subscribers. Sales teams receive information like name, email id, mobile number, gender, age, etc.

But wouldn’t it be better if a conversion score was given, for example, user A has a high potential to convert compared to user B so that the team can focus their time and efforts on user A. Or provide insights on user cohorts that consume least amount of time but have high potential to convert.

Analytics and Data science can answer these questions for the sales team. In this problem statement, a sales team has approached the analytics team to answer and solve some of its pain points.

Problem Statement (PS) – Improving the Efficiency of Sales Conversion

The below PS is an original case study assignment given by an established healthcare start-up for the senior business analyst role.

This PS falls into 0 to 1 and New Problems.

Attached is a dump of slots booked by free users over 2 months with our coaches. The leads are separated by funnel – Bot / Free-Trial and Lead Type – With/without medical condition and India/NRI. All consultations are free, and the objective is to upsell premium subscriptions. Purchase can be identified with the payment time filter. Coaches are divided into Target Classes basis their ability to sell. ‘A’ being the best in sales and ‘D’ being the worst. The data of the actual users and coaches have been masked.

  1. What’s the 3-day and 7-day conversion of different lead types split by funnels?
  2. What hours work best for connectivity and sales?
  3. Come up with insights on how best you would optimize slots, coach, and funnel for the most efficient outcome.
  4. Are there any other insights you can come up with using this data?

Load and summarise data:

health_df =pd.read_csv('https://raw.githubusercontent.com/chrisdmell/Project_DataScience/working_branch/03_healthifyme_case_study/sales_call_data_dump.csv')

print("Shape of the df")
display(health_df.shape)

print("HEAD")
display(health_df.head())

print("NULL CHECK")
display(health_df.isnull().any().sum())

print("NULL CHECK")
display(health_df.isnull().sum())

print("df INFO")
display(health_df.info())

print("DESCRIBE")
display(health_df.describe())

Python Code:

import pandas as pd
from IPython.display import display

health_df =pd.read_csv('''https://raw.githubusercontent.com/chrisdmell/Project_DataScience/working_branch/03_healthifyme_case_study/sales_call_data_dump.csv''')
print("*** Shape of the df ***")
display(health_df.shape)
print("*** HEAD ***")
display(health_df.head())
print("*** NULL CHECK ***")
display(health_df.isnull().any().sum())
print("*** df INFO ****")
display(health_df.info())
print("**** DESCRIBE ***")
display(health_df.describe())

Column Names:

Explanations of the columns had not been provided, assuming these columns’ descriptions.

  • expert_id – ID of the sales representative(categorical).
  • team_lead_id – ID of the team lead(categorical).
  • user_id –  User id(categorical).
  • India vs NRI –  Binary flag, whether Indian or not(Binary).
  • medicalconditionflag – Binary flag, any medical condition or not(Binary).
  • funnel – Source of the lead, BOT or FT(Free Trial).
  • event_type – The purpose of the call is either to book a consultation or a sales call.
  • current_status – Current status of the event.
  • handled_time – Conversation time between agent and user.
  • slot_start_time – Consultation start time.
  • booked_flag – Consultation booked or not.
  • payment_time – Subscription activation time.
  • target_class – Class of users.

Convert dates into datetime objects:

time_to_pandas_time = ["slot_start_time", "handled_time", "payment_time"]
for cols in time_to_pandas_time:
  health_df[cols] = pd.to_datetime(health_df[cols])
health_df.dtypes

What’s the 3-day and 7-day conversion of different lead types split by funnels?

health_df["max_start_handled_times"] = health_df[["handled_time", "slot_start_time"]].max(axis=1)
health_df["conversion_days"] = (health_df['payment_time'] - health_df['max_start_handled_times']).dt.days
health_df.head()
## Total base
conversion_base = health_df.groupby(["team_lead_id", "funnel"]).agg({"max_start_handled_times":"count"}).reset_index()
## in 3 days how many converted
conversion_base_3 = health_df[health_df["conversion_days"] <= 3].groupby(["team_lead_id", "funnel"]).agg({"max_start_handled_times":"count"}).reset_index()
## in 7 days how many converted
conversion_base_7 = health_df[health_df["conversion_days"] <= 7].groupby(["team_lead_id", "funnel"]).agg({"max_start_handled_times":"count"}).reset_index()
import pandas as pd
from functools import reduce
# compile the list of dataframes you want to merge
data_frames = [conversion_base, conversion_base_3, conversion_base_7]
df_merged = reduce(lambda  left,right: pd.merge(left,right,on=["team_lead_id", "funnel"],how='left'), data_frames)
df_merged["seven_day_perc"] = df_merged.max_start_handled_times/df_merged.max_start_handled_times_x
df_merged["three_day_perc"] = df_merged.max_start_handled_times_y/df_merged.max_start_handled_times_x
df_merged.head()

What hours work best for connectivity and sales?

health_df["max_start_handled_times_hour"] = health_df.max_start_handled_times.dt.hour
health_df_bt = health_df[["max_start_handled_times", "max_start_handled_times_hour","funnel", "team_lead_id" ,"India vs NRI", "medicalconditionflag","payment_time","current_status"]]
## Total base
conversion_base_H = health_df_bt.groupby(["max_start_handled_times_hour"])
                .agg({"max_start_handled_times":"count", "payment_time":"count"}).reset_index()
conversion_base_H["conversion_ratio"] = (conversion_base_H.payment_time/ conversion_base_H.max_start_handled_times)*100
conversion_base_H.sort_values(by = ["conversion_ratio"], ascending = False).head()
6 AM has the highest conversion ratio, 6.13 but only 16 people converted so the total revenue is lower compared to say 10 AM. So the best hours should be the ones where revenue soars and conversion is also. Let’s assume the unit cost of the subscription to be Rs.999.
conversion_base_H["unit_gross_amt_avg"] = 999  ## assume
conversion_base_H["gross_sales"] = conversion_base_H.payment_time*conversion_base_H.unit_gross_amt_avg
conversion_base_H.sort_values(by = ["gross_sales"], ascending = False).head()
CODE OUTPUT

When revenue is optimized, 11 AM is the best time for connectivity and sales. Generally, morning hours from 10 AM to 12 AM and evening from 5 PM to 6 PM are best, considering conversion and revenue.

Come up with insights on how best you would optimize slots, coach, and funnel for the most efficient outcome. Are there any other insights you can come up with using this data?

The introduction explains that most people use inductive learning through practice and observation. Readers can attempt questions 3 and 4 and try to solve them to improve their analytical thinking and problem-solving skills.

Some other important questions,

  1. Build a classification model and provide a conversion propensity score for each user X Agent.
  2. Should the company invest more in FT or BOT? which has higher conversion?
  3. It is observed that conversions are higher for users with a medical history. How can this be exploited to accrue higher revenues and subscriptions?
  4. Is it true that Bot has a better conversion for advanced countries, and for India, a Free trial is better?
  5. What is the cost per lead(CPL, assume missing data point if needed. For example, the cost can be assumed based on a quick google search)?
  6. Which sales lead has a higher retention rate? And why?
  7. What is the cost of acquisition(CAC) for FT and BOT(assume relevant data points if needed)?
  8. Seven-day conversion is not better; is this because of the sales agent( are these agents’ conversions worse off)? Or is this due to some customer cohort?

Solving the above questions without external help will provide immense self-confidence while solving actual interview case studies.

Our Recommendations to Data Analysts

  1. Seven-day conversion is just slightly better than three days conversions (from 31% to 33%, from 35% to 39%) so provide attractive subscription offers once 3 day time period is over. This could increase the 7-day conversion by X%(the X% can be calculated by previously run promotional campaigns).
  2. Run a KYC(know your customer survey) to understand why aren’t people converting post 3 days.
  3. Identity agents whose 3-day conversion is low and retrain them.
  4. Morning hours from 10 AM to 12 AM and evening from 5 PM to 6 PM are best, considering conversion and revenue.
  5. Readers can add more recommendations.

Conclusion

Case studies, when done right, following the steps given above, will have a positive impact on the recruiter. Recruiters aren’t looking for answers but an approach to those answers, the structure followed, the reasoning used, and business and practical knowledge. This article provides an easy-to-follow framework for data analysts using a real business case study as an example.

To summarise:

  • Use case studies as a practical resume to showcase strong fundamentals, tools, and techniques.
  • Go above and beyond, and answer a few ambiguous questions.
  • Provide all possible solutions (at least two), and don’t limit solutions to just one. Discuss the advantages and drawbacks of each solution.
  • Start with descriptive analysis, then move on to prescriptive analysis and finally provide practical recommendations that can be implemented by business teams, in this case, the sales team.
  • Highlight the outcome of the solution, for example – It will save 7000 man hours, increase revenue by 5%, etc.
  • Finally, present everything in a PPT – 1. Agenda, 2. The problem statement, 3. Supporting slide with charts and insights, 4. The solution, recommendation, and Conclusion.
  • Check out my article on – Solving Business Case Study Assignments For Data Scientists.
  • Be audacious!

Good luck! Here’s my Linkedin profile if you want to connect with me or want to help improve the article. Feel free to ping me on Topmate/Mentro; you can drop me a message with your query. I’ll be happy to be connected. Check out my other articles on data science and analytics here.

The media shown in this article is not owned by Analytics Vidhya and is used at the Author’s discretion.

Chrisx10
Chrisx10

Data scientist. Extensively using data mining, data processing algorithms, visualization, statistics, and predictive modeling to solve challenging business problems and generate insights. My responsibilities as a Data Scientist include but are not limited to developing analytical models, data cleaning, explorations, feature engineering, feature selection, modeling, building prototype, documentation of an algorithm, and insights for projects such as pricing analytics for a craft retailer, promotion analytics for a fortune 500 wholesale club, inventory management/demand forecasting for a jewelry retailer and collaborating with on-site teams to deliver highly accurate results on time.

Free Courses

Generative AI 4.7

Generative AI - A Way of Life

Explore Generative AI for beginners: create text and images, use top AI tools, learn practical skills, and ethics.
Large Language Models 4.5

Getting Started with Large Language Models

Master Large Language Models (LLMs) with this course, offering clear guidance in NLP and model training made simple.
Prompt Engineering 4.6

Building LLM Applications using Prompt Engineering

This free course guides you on building LLM apps, mastering prompt engineering, and developing chatbots with enterprise data.
LlamaIndex 4.8

Improving Real World RAG Systems: Key Challenges & Practical Solutions

Explore practical solutions, advanced retrieval strategies, and agentic RAG systems to improve context, relevance, and accuracy in AI-driven applications.
RAG systems using LlamaIndex 4.7

Microsoft Excel: Formulas & Functions

Master MS Excel for data analysis with key formulas, functions, and LookUp tools in this comprehensive course.

Recommended Articles

Responses From Readers

Clear

Write for us

Write, captivate, and earn accolades and rewards for your work
  • Reach a Global Audience
  • Get Expert Feedback
  • Build Your Brand & Audience
  • Cash In on Your Knowledge
  • Join a Thriving Community
  • Level Up Your Data Science Game
imag

We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our Privacy Policy & Cookies Policy.

Show details