How to Crack Power BI Interviews Without Experience

How a Power BI report goes from data to users:

1. Get data       - Excel, SQL database, CSV, web
2. Transform      - Power Query: clean, shape, combine
3. Model          - tables, relationships, star schema
4. Calculate      - DAX measures
5. Visualise      - report pages, visuals, slicers
6. Publish        - to a workspace in Power BI Service
7. Refresh        - schedule (gateway for on-premises data)
8. Share          - app, dashboard, permissions

Being able to say this flow clearly answers many first questions.

Import mode
+ Fast visuals (data in memory)
+ Full DAX and Power Query features
- Data only as fresh as the last refresh
- Model size limits

DirectQuery
+ Near real-time data
+ No large copy of data in the model
- Slower visuals, depends on source speed
- Some limitations in transformations

Answer pattern: "I would choose ___ because the
business needs ___ freshness and the data size is ___."

Revenue = SUMX ( Sales, Sales[Quantity] * Sales[Unit Price] )

-- Share of the selected category in all categories
Category Share % =
DIVIDE (
    [Revenue],
    CALCULATE ( [Revenue], ALL ( 'Product'[Category] ) )
)

-- Revenue only for the North region, ignoring the region slicer
North Revenue =
CALCULATE ( [Revenue], 'Customer'[Region] = "North" )

-- Rank products by revenue
Product Rank =
RANKX ( ALL ( 'Product'[Name] ), [Revenue], , DESC, DENSE )

-- Use an inactive relationship (e.g. Ship Date) for one measure
Revenue by Ship Date =
CALCULATE (
    [Revenue],
    USERELATIONSHIP ( Sales[ShipDate], 'Date'[Date] )
)

-- Count customers who bought more than once
Repeat Customers =
COUNTROWS (
    FILTER ( VALUES ( Sales[CustomerID] ), CALCULATE ( COUNTROWS ( Sales ) ) > 1 )
)

let
    Source = Csv.Document(File.Contents("C:\data\sales.csv"), [Delimiter = ",", Encoding = 65001]),
    Promoted = Table.PromoteHeaders(Source, [PromoteAllScalars = true]),
    Typed = Table.TransformColumnTypes(Promoted, {
        {"OrderDate", type date}, {"Amount", type number}, {"City", type text}}),
    Trimmed = Table.TransformColumns(Typed, {{"City", each Text.Proper(Text.Trim(_)), type text}}),
    NoBlanks = Table.SelectRows(Trimmed, each [Amount] <> null and [OrderDate] <> null),
    Deduped = Table.Distinct(NoBlanks, {"OrderID"})
in
    Deduped

Live task checklist (keep beside you)

[ ] Restate the questions the report must answer
[ ] Set correct data types first
[ ] Remove blanks / duplicates, trim text
[ ] Create or mark a Date table
[ ] Build relationships (single direction)
[ ] Write explicit measures (no implicit sums)
[ ] 3-5 visuals max, clear titles
[ ] One slicer for date, one for category
[ ] One-line insight on the page
[ ] Save and explain your steps

Week 1  Concepts: Desktop vs Service, modes, gateway,
        refresh, workspaces. Write 1-line answers.
Week 2  Modelling: star schema, relationships, Date table.
        Rebuild one project's model from scratch.
Week 3  DAX: CALCULATE, ALL, FILTER, SUMX, RANKX,
        time intelligence. Explain each aloud.
Week 4  Timed tasks: 3 messy datasets, 60 minutes each.
        Practise walking through your portfolio project.

Portfolio walkthrough (3 minutes)

1. Business question: ______________________
2. Data: sources, size, cleaning steps
3. Model: fact and dimension tables (show diagram)
4. Key measures: 2-3 DAX measures and why
5. Report: main page, how a manager would use it
6. Insight: one finding from the data
7. Improvement: what you would add next

Question                                   | Ready?
-------------------------------------------|-------
Desktop vs Service                          |
Import vs DirectQuery                       |
What is a gateway                           |
Star schema benefits                        |
Calculated column vs measure                |
CALCULATE and filter context                |
SUM vs SUMX                                 |
Row-level security                          |
Improving report performance                |
Your project walkthrough                    |