Entry-Level Data Analyst Interview Questions and Sample Answers (SQL, Excel, Case)

Entry-Level Data Analyst Interview Questions and Sample Answers (SQL, Excel, Case)

An entry-level data analyst interview, as of November 2026, commonly mixes five parts: SQL questions, Excel tasks, basic statistics, a take-home or live case, and behavioral questions about how you work and communicate. Interviewers know you have no analyst job behind you. They want to see clear thinking, careful checking and plain explanations.

We make the CourseCareers Data Analytics course and the free Career Potential Test, so we aren't a neutral source. The questions and outlines below aren't tied to our course; use them whatever way you're preparing.

Each question below comes with what the interviewer is checking and an example answer outline to adapt.

Key takeaways

  • Expect five parts: SQL, Excel, basic statistics, a case exercise and behavioral questions.
  • Know WHERE vs HAVING, inner vs left joins and COUNT(*) vs COUNT(column) cold.
  • In Excel, be ready to explain XLOOKUP and build a PivotTable from clean, single-header data.
  • For stats, explain mean vs median and correlation vs causation with a plain example.
  • In a case, lead with the answer, show your checks and state what the data can't tell you.
  • Talk through your thinking out loud; process matters as much as the final answer.

See what entry-level data work looks like

Our free Data Analytics intro course shows the daily work behind these interview questions.

Start the free data analytics intro

What's on this page

What do entry-level data analyst interviews cover?

Most of what you'll be asked falls into five areas. Here's how they line up with the questions on this page:

AreaExample questionWhat it tests
SQL"What's the difference between WHERE and HAVING?"Querying, filtering, joining and grouping data
Excel"When would you use XLOOKUP instead of VLOOKUP?"Lookups, summaries and cleaning in a spreadsheet
Statistics basics"When is the median better than the mean?"Choosing the right summary and not overclaiming
Case exercise"Revenue fell in Q3. Find out why."Working end to end, from question to recommendation
Behavioral"Explain a finding to someone who doesn't work with data."Communication, accuracy and motivation

Every company weights these differently, so read the job post closely and ask the recruiter what each round includes.

These questions test how you work, not the skill list on your résumé. For that side of hiring, see what employers look for in junior data analysts.

Which SQL questions come up for entry-level analysts?

Most entry-level interviews include SQL, either spoken or typed into a shared editor. Talk through your thinking as you write; a clear process with a small bug often beats a silent right answer. The definitions below follow the current PostgreSQL documentation (version 19, read November 2026). Other databases behave the same way on these basics, though syntax details can differ.

"What's the difference between WHERE and HAVING?"

What they're checking: Whether you know the order a grouped query runs in.

Example answer outline: WHERE filters rows before they're grouped and summed; HAVING filters the groups after the totals are calculated. The PostgreSQL docs put it this way: WHERE "controls which rows go into the aggregate computation." So a filter on a total, like regions with more than $10,000 in sales, goes in HAVING.

"What's the difference between an inner join and a left join?"

What they're checking: Whether you'll accidentally drop rows.

Example answer outline: An inner join keeps only rows that match in both tables. A left join keeps every row from the left table and fills the right table's columns with nulls where there's no match. Example use: list every customer, including those with no orders, by left-joining customers to orders.

"COUNT(*) or COUNT(column)? Which would you use?"

What they're checking: Attention to missing values, a daily problem in real data.

Example answer outline: Per the PostgreSQL docs, COUNT(*) counts every input row, while COUNT(column) counts only rows where that column isn't null. Add one more detail if you can: SUM over no rows returns null, not zero, and COALESCE can swap in a zero.

"How would you find duplicate records?"

What they're checking: Data-cleaning instinct.

Example answer outline: Group by the columns that should be unique (say, email) and keep groups where COUNT(*) is greater than 1 using HAVING. Then say what you'd do next: check whether they're true duplicates or two real records, before deleting anything.

"What's a window function, and when would you use RANK?"

What they're checking: Whether you can go one step past basic grouping. Some entry-level interviews skip this; it's a plus if you know it.

Example answer outline: A window function calculates across related rows but, unlike GROUP BY, keeps each row in the output. ROW_NUMBER numbers rows 1, 2, 3 within a group. RANK gives tied rows the same rank and leaves a gap after them; DENSE_RANK does the same with no gap. Example use: rank products by sales within each region.

What Excel questions should you prepare for?

Excel questions are usually practical: how would you look something up, summarize it or clean it. Where we cite Microsoft, the facts come from its own support pages, read November 2026.

"When would you use XLOOKUP instead of VLOOKUP?"

What they're checking: Whether you know current lookup tools and their limits.

Example answer outline: XLOOKUP can return a value from a column on either side of the one you search, while Microsoft notes VLOOKUP only looks left to right. XLOOKUP defaults to an exact match, returns #N/A when nothing matches, and has an if_not_found argument for a friendlier result. Mention one catch: Microsoft says it isn't available in Excel 2016 or 2019, so shared files may need VLOOKUP or INDEX and MATCH.

"How would you summarize 50,000 rows of sales by month and region?"

What they're checking: Whether you reach for a PivotTable and set the data up correctly first.

Example answer outline: Put the data in a clean table with one header row and one type of data per column, then build a PivotTable with month in rows, region in columns and sales in values. Microsoft notes numbers are summed by default and text fields are counted, so check the value field says Sum. Offer a second view, such as % of Grand Total, to show each region's share.

"A dataset has blank cells, extra spaces and dates stored as text. What do you do?"

What they're checking: A careful cleaning routine, and whether you keep the raw data safe.

Example answer outline: Keep an untouched copy of the raw data. Then check each column for blanks and decide whether to fill, flag or exclude them; trim stray spaces; convert text dates to real dates; and look for duplicates. Write down each change so someone else could repeat it.

"How do you check that a spreadsheet's numbers are right?"

What they're checking: Accuracy habits. One wrong total can undo trust in a whole report.

Example answer outline: Compare totals against the source system, spot-check a few rows by hand, and look for values that are impossible (negative quantities, dates in the future). Example line: "Before I send anything, I make sure my PivotTable total matches the raw data total."

Which statistics questions do junior analysts get?

Entry-level stats questions test reasoning, not formulas. Plain-English answers with an example score well.

"When is the median better than the mean?"

What they're checking: Whether you can pick the right summary for skewed data.

Example answer outline: The mean is the sum divided by the count; the median is the middle value. The NIST statistics handbook notes that extreme values in the tails distort the mean, while the median isn't affected because it's based on ranks. Example: a few very large orders would pull average order value up, so the median better shows a typical order.

"What's the difference between correlation and causation?"

What they're checking: Whether you'll overclaim in a report.

Example answer outline: Two things moving together doesn't mean one causes the other; a third factor or timing can explain both. Example: ice cream sales and sunburns rise together because of summer weather. Say how you'd test cause, such as a controlled experiment.

"You find an outlier. What do you do?"

What they're checking: Judgment rather than reflex deletion.

Example answer outline: First check whether it's an error (a typo, a test record, a unit mix-up). If it's real, keep it and report results with and without it, or use the median. Never quietly delete data that doesn't fit.

How do you handle a take-home data case?

Many employers add a case: either a take-home task due in a few days or a live exercise. What follows is a made-up example of the kind of prompt you might get, and a way to approach it.

Example prompt (made up): "Here's a spreadsheet of 12 months of orders for an online store. Revenue fell in the third quarter. Find out why and recommend what to do. Send a short write-up."

StepWhat to showExample time (4-hour budget)
1. ClarifyRestate the question; list any assumptions15 minutes
2. Check the dataRow counts, blanks, duplicates, date range30 minutes
3. Break it downRevenue = orders × average order value, then split by product, channel and region90 minutes
4. VisualizeTwo or three clear charts, each with a one-line takeaway45 minutes
5. Write it upOne page: answer first, evidence, recommendation, limits60 minutes

The times add up to 4 hours and are only an example; follow whatever limit the employer sets.

What reviewers tend to reward:

  • The answer first. Open the write-up with your finding in one sentence, such as "Q3 revenue fell mainly because repeat orders dropped," if that's what the data shows.
  • Honest limits. Say what the data can't tell you and what you'd look at next.
  • Reproducible work. Share your queries or formulas, not just the charts.
  • Your own work. If AI tools aren't allowed, don't use them. If they are, be ready to explain every step out loud.

If you present it, practice a two-sentence answer to "What would change your recommendation?" before the call.

How do you answer behavioral questions with no analyst experience?

Behavioral questions matter a lot when you have no analyst job behind you. The Bureau of Labor Statistics (BLS) has no data analyst occupation, but for market research analysts, a close relative, it lists analytical, communication and critical-thinking skills and being detail oriented as important qualities. Build stories that show them.

"Walk me through a project you're proud of."

What they're checking: Whether you can explain an analysis from question to result.

Example answer outline: The question you set out to answer, the data and where it came from, two or three steps you took (cleaning, a key query, a chart), what you found, and what you'd do differently. Portfolio projects count; so do analyses from a job that wasn't titled analyst.

"Explain a finding to someone who doesn't work with data."

What they're checking: Communication, the skill that turns analysis into a decision.

Example answer outline: Pick a real result. Lead with what it means for the listener, give one number, then one recommendation. Skip the method unless they ask. Example line: "Customers who order twice in their first month are far more likely to stay, so a follow-up offer after the first order is worth testing."

"Tell me about a time you found a mistake in your own work."

What they're checking: Honesty and attention to detail.

Example answer outline: What the error was, how you caught it, who you told and how fast, how you fixed it, and the check you added so it doesn't happen again.

Where do CourseCareers and the CPT fit?

Our Data Analytics course is online and self-paced, built for beginners preparing for a first data job. It takes about 8 to 14 weeks, self-paced. It starts with a free introduction course, then costs $499 one time or four payments of $150, with a 14-day money-back guarantee.

The honest limits: it isn't accredited and it isn't a degree, so it won't meet a posting's degree requirement. It can't sit the interview for you, and we don't publish graduate hiring rates or pay. For public pay data on nearby occupations, see our page on what BLS data says about data analyst pay.

We also make the free Career Potential Test (CPT): 55 adaptive questions across Critical Thinking, Reading, Writing and Math, up to an hour, with a percentile you can share through a results link. Its math section includes data interpretation, so it's a free way to practice timed reasoning. It's one signal, not a credential, employers may or may not use it, and we haven't published validation studies.

Who should choose something else? If you're already in a statistics, math, economics or computer science program, your coursework covers much of this, and a portfolio project may be all you need on top. If you learn well alone, free documentation like the PostgreSQL tutorial and Microsoft's Excel help pages, plus public datasets, can carry you a long way. For how SQL and Excel compare as a first skill, read SQL vs Excel for beginners.

Our recommendation: practice out loud. Write and explain five SQL queries, build one PivotTable from messy data, prepare four short stories, and do one timed practice case with a one-page write-up. If you'd like a structured path to those skills, start with our free Data Analytics intro and decide from there.

Frequently Asked Questions

What questions are asked in an entry-level data analyst interview?
Expect SQL questions (filters, joins, grouping), Excel tasks (lookups, PivotTables, cleaning), basic statistics, a case exercise and behavioral questions about past work and communication.

How do I answer data analyst interview questions with no experience?
Use portfolio projects and analyses from any job or class as your examples, and talk through your reasoning step by step. Interviewers at this level care most about clear thinking, careful checking and plain explanations.

Do entry-level data analyst interviews include a SQL test?
Often, yes, either spoken or typed in a shared editor. Practice writing filters, joins, GROUP BY with HAVING, and finding duplicates while explaining each step.

What is a data analyst take-home assignment?
It's a short project, usually a dataset and a business question, that you complete on your own and return with a write-up. Lead with your answer, show your checks and state the data's limits.

How should I explain the difference between mean and median?
The mean is the sum divided by the count, and the median is the middle value. Say the median is better when extreme values would distort the average, such as a few very large orders.

Build the skills these questions test

Try our free Data Analytics intro course to see the work up close, then decide if the full course fits.

Try the free data analytics intro

Sources