Tech hiring Data & AI
Hire data analysts who get the numbers right.
A data analyst turns the data your business already collects into answers: what happened, why, and what to do about it. Most of the work is SQL, a BI tool and careful definitions. We shortlist analysts who have been assessed on query accuracy, metric design and how clearly they explain a finding.
01 The role
What a data analyst does, and what they ship.
A data analyst answers business questions with data. They pull it from the warehouse or the production database with SQL, check that it means what everyone assumes it means, and present the result as a dashboard, a report or a short written recommendation. The output is a decision someone can make, not a chart.
The hard part is rarely the query. It is definitions: what counts as an active customer, which timestamp marks a sale, whether refunds are in or out. Two dashboards that disagree by a few per cent can cost a team weeks. A good analyst pins the definition down, documents it, and notices when a number looks too good to be true.
An analyst is not a data engineer or a data scientist. Engineers build the pipelines that deliver clean tables; scientists build predictive models and run formal experiments. Analysts sit closest to the business. In current practice they often maintain dbt models and a semantic layer as well, and use AI assistants to draft SQL that they still have to verify.
What they ship
- Dashboards for revenue, retention, operations and marketing spend
- A metrics glossary with one agreed definition per KPI
- Cohort, funnel and retention analyses
- Ad hoc investigations: why a number moved, and whether it matters
- Weekly and monthly reporting packs for leadership or investors
- A/B test read-outs with a plain recommendation
- Cleaned, documented SQL models that other analysts can reuse
02 Skills and stack
What data analysts work with, and what is optional.
The BI tool matters less than the SQL underneath it. An analyst who is strong in Power BI can learn Looker in weeks, so tell us which warehouse and BI tool you run and how strict the match needs to be.
Core
The role cannot be done without these.
- SQL: joins, window functions, CTEs, aggregation
- Metric definition and basic data modelling
- Data visualisation and dashboard design
- Spreadsheets: Excel or Google Sheets
- Descriptive statistics and sampling sense
- Data cleaning and validation
- Explaining findings in writing and in person
Tooling
Varies from team to team.
- PostgreSQL or MySQL
- BigQuery, Snowflake or Redshift
- Power BI or Tableau
- Looker or Looker Studio
- Metabase
- dbt
- Python: pandas and Jupyter
- Excel with Power Query
- Google Analytics 4, Mixpanel or Amplitude
Adjacent
Useful, not required.
- A/B test design and significance testing
- Forecasting and regression
- ETL and pipeline basics
- Product analytics event tracking
- Financial modelling
- R
03 When to hire one
Signs you need a data analyst, and signs you need someone else.
Most hiring mistakes are made before the job description is written. Read both lists before you brief us.
Hire a data analyst when
-
Nobody trusts the numbers
Finance, product and marketing each report a different revenue figure. You need someone to reconcile the sources, settle the definitions and publish one version.
-
Developers are answering data questions
Every “can you pull this for me” request interrupts an engineer. An analyst takes that queue and usually answers faster, because they learn the schema properly.
-
You have a warehouse and nobody using it
The pipelines run and the tables exist, but there are no dashboards and no regular analysis. That is analyst work, not more engineering.
-
Reports are assembled by hand from exports
The monthly pack takes days to build and breaks when a column moves. An analyst automates it and spends the saved time on the questions behind it.
-
You are about to raise money or report to a board
Investors ask for cohort retention, unit economics and growth by segment. These need to be correct, and reproducible the second time they are asked for.
Look elsewhere when
-
Your data is scattered across systems with no pipeline
An analyst can query what exists, but building reliable ingestion and a warehouse is engineering work. Without it the analyst spends most of the week exporting CSVs.
Hire data engineers -
You want predictions, recommendations or a churn model
Describing what happened is analysis. Predicting what will happen, with a model that has to be validated and maintained, is data science.
Hire data scientists -
You need a model running inside the product
Serving a model to users, with monitoring and retraining, is an engineering job well outside an analyst’s remit.
Hire machine learning engineers -
The real gap is deciding what to build
An analyst can tell you which feature is used and by whom. Choosing what to build next, and saying no to the rest, belongs to a product manager.
Hire product managers -
You need one dashboard, once
A single report or a one-off analysis does not justify a permanent hire. Scope it as a short project.
Post a freelance project
04 How we assess
Four checks, before a name reaches you.
A CV says what someone claims. We test what they can do, then send you the evidence with the shortlist.
-
Stage 1
Profile review
We look for analysis we can read: a dashboard, a report, a public notebook or a write-up of a finding. We check whether the candidate states the question, defines the metric and says how sure they are, and whether their SQL goes beyond simple selects.
-
Stage 2
Technical test
A timed test covering SQL, including joins, window functions and aggregation traps, with spreadsheet and statistics fundamentals and a practical exercise in which the candidate interprets a dataset and says what it does and does not show.
-
Stage 3
Conversation with our team
Our team asks the candidate to walk through an analysis that changed a decision: where the data came from, what they checked before trusting it and how they presented it. We listen for scepticism about their own numbers and for plain language.
-
Stage 4
Review and levelling
Reviewers score accuracy, speed, problem-solving, communication and teamwork, then assign a level. For analysts the notes record how independently the person can take a vague business question to a defensible answer, not only which tools they list.
-
Then
Your shortlist
Assessed candidates, with notes and scores.
The engine
Evalia
Assessments run on Evalia, the assessment platform we built ourselves. Across the ProDevs network, more than 100,000 assessments have been completed.
05 Seniority
Three levels. What each one can own.
Levels differ in scope more than in speed. Match the level to how much direction you can give.
Junior
Owns a report or a dashboard
- Writes correct SQL against tables and definitions someone else has set
- Builds and maintains dashboards from a clear brief
- Checks totals against a known source before sharing anything
- Needs help framing open questions and judging what is significant
Mid-level
Owns the numbers for a business area
- Takes a vague question, narrows it and returns an answer with caveats
- Agrees metric definitions with stakeholders and documents them
- Spots broken tracking or duplicated rows before they reach a report
- Presents findings to non-technical leaders without a translator
Senior
Owns how the company measures itself
- Sets the metric definitions and modelling conventions others follow
- Decides which questions deserve analysis and which are noise
- Designs experiments with product teams and reads them out honestly
- Reviews other analysts’ work and coaches them on rigour
06 Interview questions
Questions for data analysts, and what to listen for.
None of these has one right answer. The way someone answers tells you how they work.
-
Two dashboards show different monthly revenue. How do you find out which is right?
A good answer shows: A method: compare definitions first, then filters, time zones, currency handling and join fan-out, working down to row level on a small sample. Weak answers defend the dashboard they built or blame the tool.
-
Tell me about a time a join silently duplicated rows. How did you catch it?
A good answer shows: Almost every working analyst has done this. A strong answer explains one-to-many joins, the habit of checking row counts before and after, and aggregating before joining. Someone who has never seen it has not worked with messy data.
-
How would you define “active user” for our product?
A good answer shows: They ask questions before answering: which action shows value, over what window, and who will use the number. Good candidates explain the trade-off between a strict and a loose definition. A weak one says “anyone who logged in” without pausing.
-
Sign-ups rose 30% last week. What do you check before telling the CEO?
A good answer shows: Tracking changes, bots, duplicated events, a campaign, seasonality, and then whether activation further down the funnel rose too. The instinct to distrust a pleasant surprise is what you are testing.
-
When would you use a window function instead of a GROUP BY?
A good answer shows: Running totals, rankings within a group, comparisons with the previous row and deduplication with ROW_NUMBER, all while keeping row-level detail. Candidates who can describe it only in the abstract, with no example from their own work, have probably not relied on one.
-
An A/B test shows a 2% lift. Should we ship?
A good answer shows: They ask about sample size, how long the test ran, the confidence interval, and whether anyone looked at the result early. They also weigh the cost of being wrong. “It is positive, so yes” is the weak answer.
-
Show me a chart you redesigned. What was wrong with the first version?
A good answer shows: Evidence that they think about the reader: one message per chart, honest axes, fewer colours, a title that states the finding. People who treat visualisation as decoration struggle to give an example.
07 Ways to engage
Permanent, team, payroll or project. Four routes to a data analyst.
A permanent hire is the usual route, not the only one. Choose by how long the work lasts and who should employ the person.
- Placement fee, of first-year salary. One-time.
- 10–15%
- Replacement guarantee on direct hires.
- 2 weeks
- To a first shortlist, for most roles.
- 48h
-
Direct hire
A permanent employee on your contract and your payroll. One-time placement fee of 10–15% of first-year salary.
Tech hiring -
Dedicated team
A data analyst inside a team we assemble and run for you: product, design and engineering on your roadmap, for a monthly retainer.
Dedicated teams -
Outsourcing & payroll
A full-time person in a country where you have no legal entity. ProDevs employs and pays them: 20% of salary if we find the person for you, or 5% of salary, capped at $500 a month, for payroll and compliance on someone you have already chosen.
Outsourcing -
Freelance project
A defined piece of work with a clear end. You pay the agreed price per milestone through escrow, plus a 5% client fee. Free to post.
Freelancers
08 Around the role
Who data analysts depend on, and who depends on them.
A data analyst usually sits in a central data team or is embedded in a function such as product, finance or growth. Either way they work at the boundary between the people who own the data and the people who need answers from it.
- Data engineer
- Reports broken or late tables, and agrees which models belong in the warehouse.
- Product manager
- Sizes opportunities, defines success metrics before launch and reads out the results after.
- Finance and operations leads
- Reconciles reported figures with the ledger and automates recurring reports.
- Data scientist
- Supplies clean, well-defined datasets and takes over reporting once a model is live.
- Backend developer
- Agrees event tracking and schema changes so analytics does not break on release.
09 Client testimonials
Hear it first-hand, from companies that hired.
Founders, operators and executives describe working with ProDevs, in their own words.
Play video: Client testimonial: Olufemi Seyi, COO, Casafina development company
Play video: Client testimonial: Aproko Doctor, Building AwaDoc
Play video: Client testimonial: JT Liddell, Promenade CEO
10 Questions
Hiring data analysts, answered.
Something else? Book a call (opens in a new tab) and ask us directly.
-
What is the difference between a data analyst, a data scientist and a data engineer?
A data engineer builds the pipelines and warehouse that deliver clean data. A data analyst uses that data to explain what happened and why, through SQL, dashboards and reports. A data scientist builds statistical and machine-learning models to predict what will happen. Most companies need the engineer and the analyst before the scientist.
-
Do data analysts need Python, or is SQL enough?
SQL is the part that cannot be skipped; most analyst work is done there and in a BI tool. Python with pandas helps with larger clean-up jobs, statistics and automation, and is increasingly expected at mid-level and above. If your work is mainly dashboards and reporting, strong SQL matters more than Python.
-
Which BI tool should our analyst know?
The one you already use, if you have one. Power BI, Tableau, Looker and Metabase differ in detail but share the same ideas, and a capable analyst moves between them in weeks. Tell us your warehouse and BI tool in the brief and say whether experience with them is essential or merely preferred.
-
Can a data analyst work with our data if we have no warehouse?
Yes, up to a point. An analyst can query a read replica of the production database, connect a BI tool and work from spreadsheets. Once you need data from several systems joined reliably, someone has to build pipelines, and that is data engineering. Say where your data lives today so the shortlist fits your real situation.
-
How do you assess a data analyst?
In four stages: a review of past work, a timed test that includes SQL and interpreting data, a conversation with our team about an analysis the candidate has delivered, and a joint review in which a level is assigned. Assessments run on Evalia, where more than 100,000 have been completed across the ProDevs network.
-
How long does it take to hire a data analyst through ProDevs?
For most roles you see a first shortlist within 48 hours of agreeing the brief, and most companies find their hire within two to three weeks. A direct hire carries a one-time placement fee of 10–15% of first-year salary, with a two-week replacement guarantee.
Talk to ProDevs
Tell us about the role.
Tell us the level and the stack. We will tell you how quickly we can put assessed data analysts in front of you.
Project work
Need it for a project instead?
Not every job needs a full-time hire. Describe the work, get proposals from vetted freelance data analysts, and pay per milestone through escrow. Free to post.