Data Analysis Interview Questions

Saurabh K. - "Goals : Determine if the TV series should be renewed If it should be renewed, how much should Netflix be willing to pay for this series Let's assume that the goal is to maximize subscriber retention and engagement while paying a reasonable amount for the licensing costs that is justified by the value added by the series. Assumptions : The show is exclusive to Netflix for a particular region (for eg. US) It has been on the platform for an year Netflix has subscriber level data around"See full answer

Himanshu G. - "To model ROI for a product launch, the first step is to define the timeline you're targeting Example 6 months post-launch, 1 year, or even 5 years. Tip: Start with a 1-year ROI projection to estimate near-term returns, and build a 3-year projection to evaluate growth and scalability. ROI is essentially the net return over that period: Profit=Revenue (within timeline)−Total Cost (from project start) Total Cost includes both fixed and variable costs incurred since t"See full answer

Kusheta K. - "“I once worked with a dataset that had missing and inconsistent tracking. I first evaluated data quality, cleaned what was reliable, and documented assumptions. Then I collaborated with engineers to implement improved event tracking. This ensured more accurate analysis in the future and improved decision-making.”"See full answer

Sean L. - "df.loc[ isin()] is the crucial part of the solution."See full answer

Catherine T. - "We want sales to grow, in order to have a growth in revenue. And customer usage as well as it allows to see if our product lead more engagement from our users. So to be able to see this overall evolution I would make a line chart for both : Sales : with month on x-axis and sales revenue on y-axis Customer Usage : with month on x-axis and a KPI allowing to measure customer usage (nblogins or nbsessions or nbgamesplayed, ... depending on the industry) on y-axis Moreover, after knowing th"See full answer

Kevin T. - "When a stakeholder’s request is ambiguous, I start by clarifying the goal and defining what “success” looks like. I ask targeted questions to understand the business problem, the timeframe, the scope/location, and who the analysis is for. Then I confirm definitions (metrics, segments, filters), agree on the expected output format, and restate the request back to them in one sentence before I begin."See full answer

Gabriel P. - "Hi, my solution gives the exact numerical values as the proposed solution, but it doesn't pass the tests. Am I missing something, or is this a bug? def findrevenueby_city(transactions: pd.DataFrame, users: pd.DataFrame, exchange_rate: pd.DataFrame) -> pd.DataFrame: gets user city for each user id userids = users[['id', 'usercity']] and merge on transactions transactions = transactions.merge(user_ids, how='left"See full answer

Arshad P. - "Schema is wrong - id from product is mapped to id from transactions, id from product should point to product_id in transcations table"See full answer

Katherine B. - "We want to use rigorous framework for evaluating shipping a new feature — ideally an A/B test. If an A/B test is not available, we first evaluate quantitative data; we look at feature adoption metrics, time-to-use, retention and frequency of visitation. What does the business impact of the feature on conversion rates, revenue per users and LTV, and secondarily evaluate any error rates that could be occurring after the launch of the new feature. It’s important for this analysis to perform segmen"See full answer

Maryam G. - "At one of my project, I worked on a project where we needed to collect data from different sections of a large factory and deliver it to a third-party company responsible for predictive analytics on product quality and production levels. The challenge was that each department had different data types and structures, and in many cases, direct connections were restricted due to strict security policies. My responsibility was to design and implement a solution that could gather all these heterogene"See full answer

Katherine B. - "If we’re using an A/B test we have a few decision criteria that we can use to measure success. If our primary metric has been shown to be statistically significant (and our confidence interval does not cross 0), and the gaurdrail metrics that we created have not been negatively affected, we should consider shipping. If the our p-value is not significant we can still consider shipping beta if the guardrail metrics have not been negatively affected, and we weigh the opportunity cost of not shippin"See full answer

Rishabh R. - "sum of continuous subarray and keep checking if arr[i]==arr[j]. if true increase count;"See full answer