BCG Data Analyst interview guide based on two candidate reports: a pandas-heavy online assessment, then either an HR chat or a chatbot case plus a technical round with SQL, ML and Python questions.
$90K
Avg. Base Comp
$105K
Avg. Total Comp
Not reported
Typical Rounds
Not reported
Process Length
Both candidates who shared BCG data-analyst experiences start with an online assessment built around a single dataset, and neither describes an algorithm puzzle. One says it ran on CodeSignal as four tasks; the other also describes four exercises. The work is pandas-style data wrangling: cleaning, joins, filtering, null handling, encoding or mapping categories to numbers, converting ages to integers, aggregations, and a closing prediction or forecasting step. One candidate also saved the transformed result as a CSV.
After the assessment, the reports diverge. One candidate describes an HR interview that was about getting to know them and felt ordinary. The other describes an online case with a chatbot that laid out pieces of a case, asked strategic questions including graphic interpretation, and finished with a one-minute recorded recap. That same candidate later faced three SQL questions, a couple of basic ML questions and two Python questions, followed by questions on the qualification test and resume achievements, all at a fairly basic technical level. Because each of these stages appears in only one account, treat them as possible rather than guaranteed.
To prepare, rehearse a full pandas workflow on one dataset: merge two tables, filter by date, fill nulls with the mean, encode and scale, and export a CSV. Practice explaining a chart in under a minute, and be ready to describe your resume projects concisely. A refresher on basic SQL and introductory ML is reasonable too.
Synthesized from 2 candidate reports by our editorial team.
Had an interview recently?
Share your experience. Unlock the full guide.
Real interview reports from people who went through the The Boston Consulting Group process.
Share your own interview experience to unlock all reports, or subscribe for full access.
Sourced from candidate reports and verified by our team.
Topics based on recent interview experiences.
Featured question at The Boston Consulting Group
Write a function n_frequent_words that returns the top N frequent words and their frequencies, and state its run-time
| Question | |
|---|---|
| Encoding Categorical Features | |
| Over-Budget Projects | |
| Booking Regression | |
| New Partner Card | |
| Missing Housing Data | |
| Decision Tree Evaluation | |
| Analyzing Churn Behavior | |
| Retailer Data Warehouse | |
| Data Preparation for Imbalanced Data | |
| User Event Data Pipeline | |
| Interpreting Fraud Detection Trends | |
| Your Strengths and Weaknesses | |
| 2nd Highest Salary | |
| Top Three Salaries | |
| Closest SAT Scores | |
| Longest Streak Users | |
| Rolling Bank Transactions | |
| Employee Salaries | |
| Largest Salary by Department | |
| Top 3 Users | |
| Raining in Seattle | |
| First to Six | |
| Size of Joins | |
| Find the Missing Number | |
| Fair Coin | |
| Prime to N | |
| Duplicate Rows | |
| Experiment Validity | |
| Paired Products |
Synthesized from candidate reports. Individual experiences may vary.
Both candidates report an online assessment of four tasks on a single dataset. Tasks they describe include joins, filtering, null replacement, encoding or mapping categories, scaling, aggregations, converting ages to integers, and a forecasting or price-prediction step. One saved the output to CSV. It was described as practical data wrangling, not algorithms.
One candidate reports an HR interview after the assessment that was about getting to know them more and felt nothing out of the ordinary. The report does not detail specific questions, so there is little to say about its content beyond a general, conversational tone.
A different candidate reports an online case where a chatbot described pieces of a case and asked strategic questions along the way, including graphic interpretation and what questions they would ask in an interview. It ended with a one-minute recorded case review. Only one report describes this stage.
The same candidate who reported the chatbot case describes a later interview with three SQL questions, a couple of basic ML questions and two Python questions. The interviewer then followed up on the qualification test and some resume achievements. The candidate called the technical level fairly basic and the round conversational.
