CV & Interview Prep
Data Analyst
CV tips and interview questions for junior data analyst roles in South Africa. Covers Excel, SQL, Power BI, and the analytical thinking questions SA employers use to screen candidates.
Want the full study plan?
The How to Become a Data Analyst roadmap covers every skill you need to build before the interview.
Key Skills Employers Test
- Microsoft Excel (advanced functions, pivot tables, Power Query)
- SQL (SELECT, JOIN, GROUP BY, subqueries)
- Power BI (data modelling, DAX basics, dashboard design)
- Python or R for data cleaning (advantageous)
- Data visualisation principles
- Written and verbal communication of analytical findings
- Attention to detail and data quality awareness
How to Write Your CV
Must Include
- +Specific tools with a competency indication: "Microsoft Excel (advanced: VLOOKUP, pivot tables, INDEX/MATCH, macros)" rather than just "Excel"
- +SQL experience or coursework, with the specific flavour if known (T-SQL, MySQL, PostgreSQL)
- +Any BI tool experience: Power BI, Tableau, or Looker Studio
- +A portfolio link or GitHub repository showing actual projects with real datasets
- +Python or R if you have used them, even at a basic level
- +The types of analysis you have performed: reporting, forecasting, trend analysis, financial modelling
Avoid
- xListing "data analysis" as a skill without supporting it with tools and project evidence
- xClaiming Power BI or Tableau experience without a dashboard to show. Interviewers will ask to see it
- xBurying technical skills in a generic skills section at the bottom. Put them prominently, near the top
- xIgnoring soft skills entirely. Analysts who cannot communicate findings to non-technical stakeholders are a major pain point for SA hiring managers
- xA CV longer than two pages for an entry-level role
Show a project, not just coursework
The difference between a data analyst CV that gets interviews and one that does not is usually a real project. Take a public dataset from Kaggle or Statistics South Africa, clean it, analyse it, and build a Power BI dashboard or write a short findings report. One well-documented project is worth more than five listed courses.
Quantify everything
Data analysts are expected to think in numbers. A CV that contains no quantification of any kind is a red flag. "Analysed 3 months of sales data across 4 product lines to identify a 22% revenue concentration in one SKU" is the kind of language that gets attention. Even if the project was fictional, be specific.
Highlight your SQL specifically
SQL is the most commonly tested skill in SA data analyst interviews. If you have completed SQL coursework, say so clearly and specify the type: "Completed T-SQL and SQL Server Data Manipulation coursework, comfortable with SELECT, JOIN, GROUP BY, subqueries." This one line will get your CV past more screening filters than almost anything else.
Mention the SA data context where relevant
If you have worked with South African datasets, mention the source: Stats SA, National Treasury open data, or industry-specific SA data. Demonstrating familiarity with the local data landscape is a subtle but effective differentiator for SA roles.
Common Interview Questions
Write a SQL query to find the top 5 customers by total purchase value.
Why they ask this
SQL is tested in almost every SA data analyst interview, often as a practical exercise. This is one of the most common prompts.
Sample Answer
SELECT customer_id, SUM(purchase_value) AS total_value FROM orders GROUP BY customer_id ORDER BY total_value DESC LIMIT 5; I would also ask whether we want to include cancelled orders and whether "customer" means individual or account, since those assumptions affect the query.
Red flag answer
Writing the correct query but not considering edge cases or data quality questions. Good analysts question the data before querying it.
How would you explain a complex data finding to a non-technical manager?
Why they ask this
Communication is the skill SA analysts are most frequently criticised for lacking. This question screens specifically for it.
Sample Answer
I would start with the business implication, not the method. Instead of leading with "I ran a cohort analysis," I would say "Customers who make a second purchase within 14 days are three times more likely to become long-term customers." Then I would show a simple visual, ideally one chart that supports that single point. I avoid jargon and I invite questions rather than overwhelming them with all the supporting analysis upfront.
You receive a dataset and you notice there are missing values in several key columns. What do you do?
Why they ask this
Data quality handling is a practical, daily skill. Interviewers want to see a methodical approach.
Sample Answer
First I would quantify the extent of the missing data: how many rows are affected and which columns. Then I would try to understand why the data is missing. Is it a collection issue, a system error, or expected behaviour? Depending on the context and the volume, I might remove affected rows if the sample is large enough, impute with a mean or median for numeric fields where appropriate, or flag the rows and exclude them from specific calculations while keeping them in others. I would document every decision and disclose the limitation in any report I produce.
Tell me about a time you found something unexpected in a dataset.
Why they ask this
Analytical curiosity is a differentiating trait. This question surfaces whether you go beyond the brief.
Sample Answer
While analysing a fictional e-commerce sales dataset for a study project, I noticed that sales were consistently higher on Tuesdays than any other day, which was not intuitive. I investigated and found that the business sent its newsletter every Tuesday morning. That single insight suggested the newsletter was driving meaningful direct sales and warranted further analysis on its own. I flagged it in my report as a hypothesis worth testing with a controlled send-day experiment.
What is the difference between a measure and a dimension in Power BI?
Why they ask this
Power BI is widely used in SA corporate environments. This basic data modelling question screens for genuine tool knowledge.
Sample Answer
A dimension is a categorical field used to slice or filter data, such as product category, region, or date. A measure is a calculated value, typically a sum, average, count, or ratio, such as total sales, average order value, or month-on-month growth. In Power BI, measures are written in DAX and are evaluated in the context of the dimensions applied in a visual or filter.
Accelerate Your Career
Join over 10,000+ learners. Get early access to new courses, exclusive career guides, and platform updates delivered straight to your inbox.
By subscribing, you agree to our Terms of Service and Privacy Policy. Unsubscribe anytime.