Indian unicorns and high-growth startups
Roastery Coffee House Data Analyst Interview: Rounds, Questions and Preparation
For Data Analyst roles at companies like Roastery Coffee House
What the job involves
A Data Analyst at an Indian unicorn or high-growth startup works with product, marketing, and operations teams to turn raw data into decisions. Day to day you write SQL queries, build dashboards, run experiments, and present insights that shape pricing, user growth, and supply chain choices. In fast-moving consumer brands like a specialty coffee chain, you might analyse store-level sales, customer retention, and delivery partner performance.
Role and Technical Knowledge
- 1
What is the difference between a primary key and a foreign key in a sales database for a coffee chain?
Explain that a primary key uniquely identifies each row in a table like orders, while a foreign key links to another table such as customers or stores. Give a concrete example with order_id and customer_id.
- 2
How would you calculate month-over-month growth in daily active customers for a startup?
Describe grouping by month, counting distinct customers per day, averaging daily active customers per month, then using a window function or self-join to compute percentage change. Mention handling incomplete months.
- 3
Explain the difference between mean, median, and mode. When would you use median over mean for customer spend?
Define each measure. Say median is better when spend is skewed by a few high-value orders, such as corporate catering, because it gives a typical customer value.
- 4
What is a cohort analysis and why is it useful for a fast-growing consumer brand?
Explain grouping users by first purchase month and tracking retention or repeat rate over time. Mention it reveals whether newer cohorts are sticking better or worse than older ones.
Practical Scenarios and Case Questions
- 5
A coffee chain sees a 20 percent drop in online orders in one city last week. How would you investigate?
Break the problem into data checks: compare with previous weeks, split by store, channel, and time of day. Check for app issues, delivery partner availability, competitor promotions, or weather. Suggest a dashboard to monitor daily.
- 6
You are asked to design a dashboard for the operations head to track store performance. What metrics would you include?
Include daily sales, average order value, footfall, conversion rate, wastage, and staff cost per store. Mention filters by city and date range, and a simple layout with top and bottom performers.
- 7
How would you test whether a new loyalty program increases repeat purchases?
Propose an A/B test with random assignment of customers to control and treatment groups. Define success as repeat purchase rate within 30 days, calculate sample size, and check statistical significance before rolling out.
- 8
You find that 10 percent of delivery orders are cancelled. How would you reduce this?
First quantify cancellation reasons from data: long prep time, no partner, wrong address. Then suggest operational fixes like batching orders, dynamic pricing for partners, or address validation. Mention tracking improvement over time.
Behavioural and Situational Questions
- 9
Tell me about a time you used data to solve a problem in a college project or internship.
Use the STAR format. Focus on the data you collected, the tool you used, the insight you found, and the outcome. Even a small project like analysing attendance or canteen sales works.
- 10
You are given a messy dataset with missing values and duplicates. How do you handle it?
Describe a systematic approach: profile the data, decide whether to drop or impute missing values based on percentage, remove duplicates using key columns, and document every change. Mention that you would confirm with the data owner if unsure.
- 11
A senior manager asks for a report by end of day, but you realise the data is incomplete. What do you do?
Show ownership: communicate early about the gap, share preliminary findings with clear caveats, and propose a revised timeline. Highlight that you would not hide the issue.
- 12
Why do you want to work as a data analyst at a high-growth startup rather than a large corporate?
Talk about the chance to wear multiple hats, see direct impact of your analysis, and learn quickly. Mention comfort with ambiguity and fast iteration. Avoid saying you dislike big companies.
HR and Culture Fit
- 13
What do you know about our company and how do you think data helps our business?
Research the company's model, such as a specialty coffee chain with online and offline sales. Suggest data uses like demand forecasting, personalised offers, and store expansion decisions.
- 14
How do you prioritise multiple data requests from different teams?
Explain that you assess impact and urgency, clarify the decision the request supports, and communicate trade-offs. Mention using a simple framework like effort versus impact.
- 15
Where do you see yourself in three years as a data analyst?
Show a growth path: from analyst to senior analyst or analytics lead, deepening skills in experimentation or machine learning. Tie it to the company's growth stage.
- 16
What is your expected salary for this role?
Give a range based on market research for fresher data analysts in Indian startups, and say you are open to discussion based on overall learning and responsibilities. Avoid quoting a single rigid number.
Coding round
Data Analyst interviews usually include a coding round. Practise planning the solution out loud, then review code the AI writes from your plan, in Python, Java, C++, C or JavaScript.
Ready to try it for real?
Take a timed test or talk through a mock interview for this role, then see your scores.
This is practice material for Data Analyst roles, written with AI. It is not a list of questions Roastery Coffee House or any other company has asked.