← Professional profile

أحمد محمود سويد

عربي
Training profile

Business Analysis & Dashboard Design with Power BI

A 28-hour hands-on programme delivered by Ahmed Mahmoud Swid, Business & Data Analyst at Riyadh Municipality and a PMI-certified CAPM. It starts from a raw file with real defects and ends with a published, secured report that refreshes with a single click.

+150

Trainees

28

Training hours

12

Hands-on sessions

4

Weeks

▶01

From the training room

A week-long Power BI workshop for Riyadh Municipality staff — West and North sectors. Clips carrying the Municipality's identity are from its own internal-communications coverage.

Official

Power BI workshop — West sector

11–15 January 2026 · official Riyadh Municipality coverage

Official

Power BI workshop — North sector

Official coverage · Riyadh Municipality

Teaching on the live screen

Hands-on session

Working through it with the trainees

Hands-on session

The training room

Week-long workshop

Certificates awarded

Closing session

Sample output — interactive sales dashboard

Live report recording · training data

◆02

Audience and delivery requirements

Who it is for
Analysts, engineers, operations and monitoring staff, and report writers — anyone living in Excel who needs to move to a data model that refreshes with one click.
Prerequisites
A Windows machine, Power BI Desktop (free), and working Excel knowledge. No programming experience required.
Format
12 sessions across four weeks (3 per week), a one-week intensive workshop, or one-to-one coaching.
Session length
Two to two and a half hours, including 15 minutes reviewing the previous session's assignment.
Group size
12 to 16 trainees. Every trainee works on their own machine, so a larger group loses individual follow-up.
Language
Arabic, with technical terms kept in English as they appear in the tool and its documentation.
Facilities
A room with machines or personal laptops, a display screen, and internet access for publishing to Power BI Service in the final session.
❖03

Curriculum

Four weeks. Each one ends with an output the trainee can see before moving on.

1

From mess to a clean table

7 hours

S1·2 h

Where Power BI sits, and the data cycle

  • The data cycle: generated → stored → analysed → consumed → raises new questions
  • The full path: sources → Power Query → model → visuals → publish
  • Excel and Power BI: each has its strength; neither replaces the other
  • Licensing: what the training needs, and what sharing needs later

Lab · Open the tool, move between the three views, and name them

S2·2.5 h

Power Query — import and clean

  • The editor: queries · applied steps · preview
  • Transform Data, not Load — the first decision, and the most commonly wrong one
  • A file with title rows: remove blanks → remove top rows → promote headers
  • Review types after every promotion; the Any type is not allowed
  • Case sensitivity: normalise before filtering, never after
  • Arabic is harder: alef forms · taa marbuta · yaa · double spaces

Lab · Clean a raw records file until it is fit to load

S3·2.5 h

Power Query — combine and automate

  • Split by Delimiter — a compound column and its hidden spaces
  • Merge with all six join kinds, reading the match indicator diagnostically
  • Anti Join — "which record has no reference?", the first auditing tool
  • Append across two files with mismatched columns
  • Folder connector: a new monthly file flows in with no edits
  • Disable load, document the steps, map the dependencies

Lab · Build the full query set and run one refresh that cleans everything

Week output · A file that refreshes with one click, where it used to be two hours of manual work every month.

2

The model

7 hours

S4·2.5 h

Dimensional modelling and relationships

  • Why one wide table is not enough
  • The star schema: fact tables and dimension tables
  • A relationship: key · direction · cardinality
  • Bi-directional relationships: when they are needed, why they are avoided by default
  • Inactive relationships and when to activate them
  • The three-way trade-off: simplicity ⟷ performance ⟷ maintainability

Lab · Connect the fact table to its dimensions and review filter direction

S5·2 h

Date table and hierarchies

  • Turn off auto date/time first — it creates a hidden table per date column
  • Build an explicit date table and mark it as such
  • Hierarchies: top level → lowest, and drilling between them
  • Column properties: format · category · summarisation · sort by column
  • What is not used gets hidden, not deleted

Lab · A complete date table and a working geographic hierarchy

S6·2.5 h

DAX — measures and calculated columns

  • The core difference: a column computes row by row and is stored; a measure computes at display time
  • When a column and when a measure — a question that recurs for a career
  • The basics: SUM · COUNTROWS · DISTINCTCOUNT · DIVIDE
  • DIVIDE, not the division operator — dividing by zero breaks the report
  • Logic: IF · SWITCH
  • Organising measures in a dedicated table

Lab · Eight core measures, written and reviewed

Week output · A clean model that answers questions the raw table could not.

3

The heart of the tool

7 hours

S7·2.5 hPivotal session

Filter context

  • Why does the same measure return a different number in a different visual?
  • Row context versus filter context
  • How a visual imposes its context: axis · slicer · filter · row
  • The pivotal exercise: one measure shown in five places returning five numbers, then explaining each
  • No move to session 8 until the trainee explains the difference unaided

Lab · Diagnose three "wrong" numbers and name the cause of each

S8·2.5 h

CALCULATE and its family

  • CALCULATE — the function that modifies context, and the core of all DAX
  • FILTER: when it is required, and when it is an expensive luxury
  • ALL · ALLEXCEPT · REMOVEFILTERS for removing filters
  • Ratios: part of group, and group of total
  • Quick measures are used — then the generated code is read and understood

Lab · Ratios and rankings that react correctly to slicers

S9·2 h

Time intelligence

  • TOTALYTD · SAMEPERIODLASTYEAR · DATEADD · PREVIOUSMONTH
  • None of it works without a marked date table — the payoff from session 5
  • The current month is incomplete, so it is not compared to a full month
  • Trend is read over a period average, not month against month
  • Metric direction: a rising count can mean falling performance, and inverting it reverses every judgement

Lab · Month-on-month and year-on-year comparison, read correctly

Week output · The trainee writes their own measures — and diagnoses their own mistakes.

4

Presentation and delivery

7 hours

S10·2.5 h

Choosing the right visual

  • The selection rule: comparison → bars · change over time → line · composition → rarely a pie
  • Cards and KPIs, and when they replace a chart entirely
  • Slicers: their kinds and their effect on context
  • Conditional formatting against a written rule, not taste
  • Theme, background, and page grid
  • A visual serves a question; a visual without one is deleted

Lab · A complete KPI page

S11·2 h

Interaction and navigation

  • Edit interactions between visuals — the most overlooked feature
  • Drill down, drill up, and drill through
  • Custom page tooltips
  • Selection pane, bookmarks, and buttons
  • The filter pane and its three levels

Lab · A three-page report with working navigation

S12·2.5 h

Publishing, governance, and the capstone

  • Publishing to the service · workspaces · sharing
  • Dashboard versus report: the difference, and when each applies
  • Alerts and scheduled refresh
  • Row-level security: each party sees only its own scope — and it gets tested
  • Decide what enters the model before people build on it; deleting later breaks their reports

Lab · Capstone: from the raw file to a published, secured report

Week output · A published, secured report the trainee presents in fifteen minutes.

✦04

Training method

Five rules drawn from training more than 150 staff.

  1. 1

    No slide gets read aloud

    Teaching happens on the live screen, and the trainee works alongside you rather than watching.

  2. 2

    Mistakes are made on purpose

    Then diagnosed. Diagnosing an error fixes the lesson better than the correct path does.

  3. 3

    "Why" before "how"

    A professional trainee will not accept a step without a reason — that is an asset, not an obstacle.

  4. 4

    English term, Arabic meaning

    Because after the course they will read the documentation and work with English tools.

  5. 5

    The stop session

    If filter context does not land, session 7 is repeated and the course does not move on.

❂05

Training data

Six purpose-built files, each carrying one deliberate defect that serves one lesson. The principle: a clean file teaches nothing — the trainee learns from the mess.

  • Raw records (CSV)

    Deliberate defect · Three title rows, columns with no names

    Session 2 · cleaning

  • Reference table (XLSX)

    Deliberate defect · A table inside a sheet with side notes

    Session 3 · table, not sheet

  • Scopes (XLSX)

    Deliberate defect · Two fields in one column, with padding

    Session 3 · split and trim

  • Parties (CSV)

    Deliberate defect · The same name in two spellings

    Session 3 · normalise before joining

  • Monthly files (6)

    Deliberate defect · Identical structure, renewed monthly

    Session 3 · folder connector

  • Prior-year records

    Deliberate defect · Two columns differ from its sibling

    Session 3 · mismatched append

And defects seeded through the rows: text dates · 1/0 codes needing translation · inconsistent casing · double Arabic spaces · records with no reference for the Anti Join lesson · outliers in handling time.

◈06

Assessment

Assignment after every session
30 to 45 minutes, reviewed in the first 15 minutes of the next session.
Three checkpoints
At the end of each of the first three weeks.
A mandatory gate
Session 7: no progress until the trainee explains filter context in their own words.
Capstone project
Raw file → published, secured report, presented by the trainee in fifteen minutes.

What the trainee leaves with

  • ✓A complete .pbix file they built themselves from a raw source
  • ✓A report published to Power BI Service with permissions configured
  • ✓A written glossary: the English term and its agreed Arabic equivalent
  • ✓A copy of the training data to repeat the exercises after the course
  • ✓A pre-publish checklist for any future report
▤07

Prepared material

Typeset, print-ready handbooks derived from a single source script, so the terminology never diverges between them.

▤

Curriculum plan and language standard

PDF · 7 pages · Arabic

Open file ↗▤

Part One — Introduction to Power BI

PDF · 22 pages · Arabic

Open file ↗▤

Part Two — Your first complete report

PDF · 23 pages · Arabic

Open file ↗

To arrange a course

The programme adapts to the centre's schedule, group size, and starting level, and the training data can be rebuilt around the trainees' own domain on request.

WhatsAppEmailFull professional profile