JPJobPrepfull-stack interview
RoadmapsJS CompilerStar on GitHub
Career Roadmaps · BI Developer

Career roadmap

BI Developer

Build the reporting layer an organisation runs on, from data model to governed, performant dashboards.

Time
4-7 months part-time
Entry bar
Analytical background. Very accessible from finance or operations roles.
Stages
5 · 25 topics
0/25 studied0%
All roadmaps

Before you start BI Developer

  • SQL fundamentals
  • Spreadsheet modelling experience
  • Understanding of business reporting needs

Data foundations

4-5 weeks · 0/5 topics

A dashboard is only as good as the model beneath it.

  1. Reporting SQL has its own patterns: date spines, aggregation and reconciliation.

    • Joins, aggregation and grouping
    • Window functions for period comparison
    • Date dimensions and calendars
    • Reconciling totals to source systems

  2. Star schemas exist because they make reporting fast and comprehensible.

    • Facts, dimensions and grain
    • Star versus snowflake trade-offs
    • Role-playing dimensions
    • Handling many-to-many relationships

  3. Getting data into shape is usually most of a BI project.

    • Power Query and dataflows
    • Incremental refresh design
    • Handling source system quirks
    • Staging versus direct query

  4. Knowing what the warehouse can do stops you rebuilding it in the BI tool.

    • Warehouse versus data mart
    • Views, materialised views and tables
    • Push-down of aggregation
    • Working with the data team

  5. One wrong figure in a board pack undoes a year of credibility.

    • Validation against source
    • Handling late and restated data
    • Null and edge case behaviour
    • Audit trails for reported figures

BuildBuild a star schema from a transactional source and a first report on top of it.

Building reports

5-6 weeks · 0/5 topics

Tool depth: the calculation language is where BI developers are actually tested.

  1. DAX or LookML fluency is the technical core of the role.

    • Measures versus calculated columns
    • Filter context and context transition
    • CALCULATE and its patterns
    • Debugging unexpected results

  2. Every business report is a comparison against a previous period.

    • Year to date and rolling periods
    • Prior period and prior year comparison
    • Fiscal calendars
    • Handling incomplete current periods

  3. Clarity beats decoration, and interviewers will critique your layout.

    • Layout and visual hierarchy
    • Chart selection for the question
    • Consistent formatting and theming
    • Accessibility and colour choices

  4. Letting users answer their own follow-up questions.

    • Slicers, filters and bookmarks
    • Drill-through and drill-down
    • Parameters and what-if analysis
    • Tooltips that add information

  5. Getting this wrong is a data breach, not a bug.

    • Static and dynamic RLS
    • Testing security as different users
    • Object-level security
    • Common RLS performance pitfalls

BuildA multi-page report with time intelligence, drill-through and row-level security.

Performance and scale

3-5 weeks · 0/5 topics

Slow reports are abandoned reports.

  1. A signature BI interview topic, and the most common real complaint.

    • Model size and column cardinality
    • Measure efficiency and iterators
    • Reducing visual count per page
    • Aggregation tables

  2. Import, DirectQuery or hybrid, chosen for a reason rather than by default.

    • Import versus DirectQuery trade-offs
    • Composite models and aggregations
    • Refresh windows and incremental refresh
    • Real-time reporting options

  3. Measuring where the time goes rather than guessing.

    • Performance analyser and query traces
    • DAX Studio and VertiPaq analysis
    • Identifying the expensive measure
    • Warehouse-side query tuning

  4. Techniques that only matter once the model outgrows memory.

    • Partitioning strategy
    • Aggregation awareness
    • Reducing model footprint
    • Archiving historical data

  5. Knowing which reports are used tells you what to maintain and what to retire.

    • Usage metrics and adoption tracking
    • Refresh failure alerting
    • Capacity monitoring
    • Retiring unused content

BuildTake a slow report from thirty seconds to under three, and document what caused the difference.

Governance and delivery

3-4 weeks · 0/5 topics

Enterprise BI is as much about control and process as about charts.

  1. Uncontrolled report sprawl is the standard enterprise BI failure.

    • Workspace and content lifecycle
    • Certified and promoted datasets
    • Naming and ownership standards
    • Managing self-service sprawl

  2. BI is software, and increasingly treated as such.

    • Development, test and production workspaces
    • Deployment pipelines
    • Version control for BI artefacts
    • Change review process

  3. Most failed BI projects failed at the requirements stage.

    • Interviewing stakeholders effectively
    • Defining metrics unambiguously
    • Managing scope and expectations
    • Prototyping before building

  4. A report nobody understands is a report nobody uses.

    • User training and documentation
    • Report walkthroughs
    • Handling change requests
    • Building analyst self-sufficiency

  5. Getting reports to where people already work.

    • Embedded analytics
    • Subscriptions and alerts
    • Mobile report design
    • Export and paginated reporting

BuildSet up a governed workspace structure with deployment pipelines and documented ownership.

Interview preparation

2-4 weeks · 0/5 topics

BI interviews are tool-specific, practical, and often include a build exercise.

  1. Deep questions on the specific platform the employer uses.

    • DAX filter context questions
    • Model relationship troubleshooting
    • Storage mode decisions
    • Security implementation

  2. A dataset and a brief, usually with a short deadline.

    • Scoping to the time available
    • Model first, visuals second
    • Documenting assumptions
    • Presenting the result

  3. Still tested, since most BI work starts with a query.

    • Aggregation and joins
    • Window functions
    • Date logic
    • Query optimisation basics

  4. Users say the numbers are wrong. What do you do.

    • Reconciliation methodology
    • Distinguishing model from source errors
    • Managing stakeholder confidence
    • Preventing recurrence

  5. Published reports with real data are the accepted evidence.

    • A public interactive report
    • Documented data model
    • A performance optimisation case study
    • Certification in your target tool

BuildA public portfolio report with a documented model and a written design rationale.

BI Developer tools on your CV

  • Power BI
  • Tableau
  • Looker
  • DAX
  • SQL
  • Power Query
  • DAX Studio
  • SSAS / Tabular

What BI Developer employers ask to see

  • A published interactive report with a documented model
  • A performance optimisation case study with timings
  • A governed workspace structure you designed
  • A vendor certification in your target tool

Enterprise demand is steady and geographically broad, especially in finance, healthcare, manufacturing and the public sector. Less competitive than data science with comparable stability.

Content last reviewed 2026-08-31. Guidance only — no institute or paid placement is endorsed anywhere in this book.