How to find a data analyst.

Every data analyst CV lists SQL, and the word covers everything from writing a SELECT statement against a prepared view to optimising complex queries with window functions. The CV cannot tell you which, and most interview processes never establish it.

A second mismatch is more consequential. Most advertised analyst roles involve maintaining recurring reports rather than investigating open questions. Both are real jobs, but candidates who want investigation leave reporting roles quickly. Describing the actual balance costs nothing and prevents a predictable departure six months in.

Job titles worth searching

Grouped by what the person actually does, because searching all45 at once produces a result set you cannot triage. Decide which group you need first — that decision does more for the search than any string below.

Where data analysts actually are

Portfolio platforms make this role unusually assessable before contact. Tableau Public hosts published dashboards that show analytical approach and design judgement directly, and GitHub carries analysis projects and dbt models. For a role where CVs are notoriously hard to distinguish, this is a genuine advantage that few recruiters use.

The visualisation community runs open weekly challenges — Makeover Monday and Workout Wednesday among them — with public submissions. Participants are demonstrably practising beyond their day job, which correlates well with the curiosity the role rewards, and almost nobody sources from them.

The most reliable underused pool is career changers from operational roles within the same industry. An analyst who spent years in the business can tell when a result is implausible; one who cannot will report it confidently. Because their history is not in analytics, keyword screening rejects them systematically, which is exactly why they remain available.

Boolean search strings

Written to be pasted as-is. Each one is built around an intent rather than a platform, since the useful question is what you are trying to find, not which site you happen to be on.

site:linkedin.com/in/ ("data analyst" OR "BI analyst" OR "insights analyst") (SQL OR "Power BI" OR Tableau OR dbt) "{city}"

("data analyst" OR "BI analyst") ("window function" OR "CTE" OR "query optimization" OR "stored procedure") -jobs -course

(site:github.com OR site:public.tableau.com OR site:novypro.com) ("dashboard" OR "analysis" OR "portfolio") "{topic}" -tutorial

("analytics engineer" OR "dbt") ("data modelling" OR "semantic layer" OR "testing" OR "documentation") -jobs -course

("analyst") ("attribution" OR "cohort" OR "churn" OR "LTV" OR "variance analysis" OR "demand forecasting") -jobs -course

("A/B test" OR "statistical significance" OR "causal inference" OR "confidence interval") ("analyst" OR "experimentation") -jobs -course

(site:kaggle.com OR "Makeover Monday" OR "Workout Wednesday") ("dashboard" OR "visualization" OR "analysis") -jobs

("transitioning to data" OR "career change" OR "bootcamp graduate") ("analyst" OR "analytics") "{industry}" -jobs

Mistakes that cost the most time

1. Advertising an analysis role that is actually reporting

   Most advertised analyst roles involve maintaining recurring reports and dashboards rather than investigating open questions. Both are legitimate, but candidates who want investigative work leave reporting roles quickly, and candidates who prefer structured delivery struggle with ambiguity. Describing the actual balance in the advert prevents a predictable early departure.

2. Not testing SQL depth properly

   Nearly every analyst CV lists SQL, and the range of what that means is enormous — from writing SELECT statements against a prepared view to optimising complex queries with window functions and CTEs. A short practical exercise reveals more in fifteen minutes than any amount of CV review, and the difference materially affects what the person can do independently.

3. Screening on BI platform rather than thinking

   Power BI, Tableau, and Looker are learnable in weeks by someone who understands data modelling and can write SQL. The analytical judgement — knowing which question to ask, spotting when a number is wrong, communicating a finding — takes years. Requiring an exact platform match filters for the easily acquired skill and against the hard one.

4. Ignoring the domain knowledge advantage

   An analyst who understands the business can tell when a result is implausible; one who does not will report it confidently. Career changers from operational roles in the same industry bring that judgement, and because their history is not in analytics, keyword screening rejects them systematically. This is one of the most reliable sources of good analysts.

5. Underestimating stakeholder communication

   Much of an analyst's value lies in explaining a finding to people who will not read the methodology, and in pushing back when a stakeholder asks for a number that will mislead them. This is rarely tested and frequently the reason a technically capable analyst underperforms. Ask how they handled a request they thought was wrong.

6. Confusing analysts with data scientists

   Advertising a data scientist role when the work is dashboarding and reporting attracts candidates who will be dissatisfied, and often costs more. The reverse — hiring an analyst into a role that genuinely needs statistical modelling — leaves the work undone. The honest title attracts the person who wants that job.

Common questions

What job titles should I search for when hiring a data analyst?

Core terms include Data Analyst, BI Analyst, Reporting Analyst, Insights Analyst, and MI Analyst, which are largely interchangeable and set by employer convention. Function-specific titles such as Marketing Analyst, Finance Analyst, and Operations Analyst identify domain context that often matters more than tooling. Platform titles — Power BI Developer, Tableau Developer, Looker Analyst — are worth searching where an organisation has heavy investment in one. For roles needing inference rather than description, Quantitative Analyst, Experimentation Analyst, and Product Analyst are more accurate.

What is the difference between a data analyst and a data scientist?

A data analyst answers questions about what happened and why, working primarily with SQL and BI tools to produce reporting and analysis for business stakeholders. A data scientist typically builds predictive models or runs statistical inference, requiring deeper mathematical grounding and usually programming in Python or R. The boundary blurs, and many organisations use the titles loosely, but the practical distinction matters for hiring: advertising a data scientist role when the work is dashboarding attracts dissatisfied candidates and costs more, while hiring an analyst for genuine modelling work leaves it undone.

How do I assess whether an analyst's SQL is actually good?

With a short practical exercise, because the CV cannot tell you. Nearly every analyst lists SQL, and the range spans writing simple SELECT statements against a prepared view through to optimising complex queries with window functions, CTEs, and an understanding of execution plans. A fifteen-minute task involving a join, an aggregation, and a window function reveals the level immediately. This matters because it determines whether the analyst can work independently or needs a data engineer to prepare everything first.

Does BI platform experience need to match exactly?

Rarely, and requiring it filters for the wrong thing. Power BI, Tableau, Looker, and Qlik are all learnable within weeks by someone who understands data modelling and writes competent SQL, because the underlying concepts are shared. What takes years to develop is analytical judgement — knowing which question to ask, recognising when a number is implausible, and communicating a finding to people who will not read the methodology. Screening on exact platform match selects for the easily acquired skill and against the difficult one.

Where can I find data analysts outside LinkedIn?

Portfolio platforms are unusually useful for this role. Tableau Public hosts published dashboards showing actual analytical and design judgement, and the visualisation community runs open weekly challenges such as Makeover Monday and Workout Wednesday whose participants are demonstrably practising beyond their day job. GitHub carries analysis projects and dbt work. An underused pool is career changers from operational roles in the same industry — they bring business judgement that pure analysts lack, and keyword screening rejects them automatically.