Niyo Data Analyst Interview: Rounds, Questions and Preparation

What the job involves

Data analysts at Indian unicorns and high-growth startups like Niyo spend their day pulling data from multiple sources, writing SQL queries, building dashboards, and translating numbers into actionable insights for product, marketing, and operations teams. They work closely with business stakeholders to define metrics, run experiments, and track KPIs in fast-changing environments where speed and accuracy matter. Expect to juggle ad-hoc requests, automate reports, and present findings that directly influence decisions.

Role and Sector Knowledge

1. 1

   What does a typical day look like for a data analyst at a high-growth fintech startup like Niyo?

   Mention morning stand-ups, ad-hoc data pulls, dashboard updates, and collaborating with product or marketing on metrics. Show awareness of fast-paced, iterative work.

2. 2

   How would you define and track a key metric like 'monthly active users' for a neo-banking app?

   Explain the definition clearly (e.g., unique users who perform at least one transaction or login in a month), data sources needed, and how to avoid double counting.

3. 3

   Why do you want to work as a data analyst in the Indian startup ecosystem rather than a large MNC?

   Highlight the opportunity to wear multiple hats, see direct impact of your work, and learn quickly. Avoid saying MNCs are boring; focus on your preference for ownership.

4. 4

   What are some common data quality issues you might face in a startup and how would you handle them?

   Give examples like missing values, duplicate entries, or inconsistent timestamps. Suggest validation checks, documentation, and communicating with data engineers.

Technical Skills and Tools

1. 5

   Write a SQL query to find the top 5 users by number of transactions in the last 30 days.

   Use JOIN between users and transactions, filter by date, GROUP BY user, ORDER BY COUNT DESC, and LIMIT 5. Mention handling of time zones if relevant.

2. 6

   How would you use Python to clean a dataset with missing values and outliers?

   Talk about pandas functions like dropna, fillna, and describe. For outliers, mention IQR or z-score and decide whether to cap, remove, or keep based on context.

3. 7

   What is the difference between a LEFT JOIN and an INNER JOIN? Give a scenario where you would use each.

   Explain that LEFT JOIN keeps all rows from the left table, INNER JOIN keeps only matching rows. Use example: LEFT JOIN to find users without transactions, INNER JOIN for completed transactions.

4. 8

   How would you design a dashboard to monitor daily sign-ups and activation rate for a fintech app?

   Mention choosing the right chart types, filters for date and channel, and defining activation clearly (e.g., first transaction). Keep it simple and actionable for stakeholders.

Practical Scenarios and Case Studies

1. 9

   A product manager asks you to find why user retention dropped last month. How would you approach this?

   Start by confirming the metric definition, break down by cohorts, channels, and geographies. Check for data issues, then look at external factors like app updates or competition.

2. 10

   You are asked to measure the success of a new feature that offers cashback on first transaction. What metrics would you track?

   Suggest activation rate, first transaction completion, repeat transaction rate, and incremental lift. Mention A/B testing if a control group exists.

3. 11

   How would you estimate the number of daily active users for a new savings product in India?

   Use a bottom-up approach: target user base, adoption rate, and frequency of use. Show structured thinking and state assumptions clearly.

4. 12

   If your dashboard shows a sudden spike in errors, how would you investigate and communicate it?

   Check data pipeline, source logs, and recent code changes. Communicate quickly with engineering and stakeholders, and provide a timeline for resolution.

Behavioural and HR Questions

1. 13

   Tell me about a time you had to explain a complex data finding to a non-technical person.

   Use STAR method: situation, task, action, result. Emphasize simplifying language, using visuals, and checking for understanding.

2. 14

   Describe a situation where you had to meet a tight deadline with incomplete data. What did you do?

   Show prioritization, communicating assumptions, and delivering a 'good enough' answer with caveats. Highlight learning and follow-up.

3. 15

   How do you handle feedback or criticism about your analysis from a senior stakeholder?

   Demonstrate openness, asking clarifying questions, and validating their concerns with data. Mention separating ego from work and improving iteratively.

4. 16

   Where do you see yourself in three years as a data analyst in a startup?

   Align with growth: deeper technical skills, owning a domain, or mentoring juniors. Show enthusiasm for the company's mission and the startup journey.