What to assess before interviewing a data analyst

You open a data analyst candidate’s resume and there they are: SQL, Python, Power BI, maybe some dbt or Looker. An impeccable list. But that list doesn’t tell you the one thing that truly matters: does this person reason well with data, or do they just know how to run queries? Can they explain to a commercial manager why conversion dropped, or do they stare at the dashboard without drawing a conclusion?

In LATAM, where many data teams are small and the analyst talks directly to the business, that gap is costly. You hire someone who masters the tools and discover too late that they don’t frame good questions, don’t catch a suspicious data point, or can’t get the rest of the team to understand what they found. That’s why it’s worth assessing before you interview: arriving at the conversation knowing where the real doubts are.

Why the resume isn’t enough for this role

A data analyst’s resume is deceptively strong. Tools are easy to list and almost all of them can be learned with a course. What’s hard to learn —and what the resume never shows— is judgment: knowing which metric to look at, distrusting an average that hides two different populations, or deciding when a finding warrants an alert and when it’s noise.

You won’t see in the resume how they communicate either. An analyst who finds something valuable but can’t tell it in business language ends up being a generator of reports nobody reads. The resume gives you inputs; you need evidence beyond the resume to know how they think and how they translate data into decisions.

Which signals to observe before the interview

Rather than confirming tools, it’s worth looking at signals of reasoning and craft. Three that tend to make the difference:

• Applied quantitative reasoning: not abstract operations, but how they interpret a trend, a proportion or an anomaly with business context.
• Structured thinking: how they break down an ambiguous problem (“why did sales drop?”) into hypotheses verifiable with data.
• Communication of findings: whether what they find can be explained clearly to someone who isn’t technical.

These signals are comparable across candidates when you observe them with the same criteria, not according to the impression of whoever interviews that day. That’s where a prior assessment gives you a common starting point for the whole panel.

How to combine competencies by role

A data analyst isn’t defined by a single competency. The useful profile combines numerical reasoning, the ability to structure problems and communication, with weights that change by role: in a business-facing role, communication weighs more; in one close to data engineering, technical rigor does. Combining competencies by role avoids the mistake of assessing everyone with the same yardstick.

What to look at in the report

When you review an assessment report, don’t look for a number that decides for you: look for inputs to better prepare the interview. Look at the role fit signal as a starting point, but go down to the competency-by-competency detail. If someone shows good reasoning but weaker communication, you already know where to focus the conversation.

Also review the process’s integrity controls, which help you read the results with context before inviting someone to interview. And remember: the report supports the decision, the team keeps it. Kokoro gives you backing to decide who to interview and with what questions; the final choice remains yours. If you work in a data-driven environment, in solutions for technology you’ll see how this fits into technical profiles.

Evidence-based interview questions

The interview pays off far more when you arrive having read the assessment. Instead of re-asking what you already know, you dig in where the report showed doubts. Some lines that work for a data analyst:

• “Tell me about a time the data contradicted what the team expected. What did you do?”
• “You have a sales report with an odd spike in one region. Where do you start?”
• “Explain a technical finding of yours to me as if I were from finance.”
• “How do you validate that a query or a calculation is right before presenting it?”

The answers, contrasted with what you saw in the assessment, give you common criteria to compare candidates without depending on the chemistry of the moment. In the resources library you’ll find more guides by role to build these conversations.

In short: for a data analyst, stop counting tools in the resume and start observing how they reason, how they look after rigor and how they communicate. Assess before you interview, arrive with the doubts mapped out and use the interview to confirm what the evidence already hinted at. This way you decide with backing who to interview, and the team keeps the final decision. You can start by looking at the competency combination for a data analyst.