Duration 3 days – 21 hrs
Overview
A hands-on program that teaches participants how to clean messy data, reshape it into analysis-ready tables, and turn it into clear summary views and interactive dashboards that inform decisions. Using either Power BI (Power Query + DAX) or Tableau (Tableau Prep + LOD/Calcs), learners will practice end-to-end: from importing raw files, fixing quality issues, modeling the data, building summaries/KPIs, and presenting insights with best-practice visuals and storytelling.
Objectives
- Identify common data quality issues (duplicates, missing values, type errors, inconsistent categories).
- Clean and reshape data using Power Query or Tableau Prep (split/merge, pivot/unpivot, fuzzy match, joins).
- Build a simple star schema (facts & dimensions) and manage relationships.
- Create summary tables and KPIs (YoY, MoM, contribution, rank, variance).
- Write essential DAX (Power BI) or Tableau calculations/LOD to enable robust measures.
- Apply visualization best practices (chart selection, labeling, color, layout, accessibility).
- Design interactive dashboards with filters, drill-throughs, tooltips, and actions.
- Publish and share dashboards securely; set up refresh schedules; apply basic governance.
- Tell a concise, decision-oriented data story aligned to business questions.
- Validate results and document transformation steps for reproducibility.
Audience
- Business
- Data Analysts, MIS
- Reporting teams, Operations & Finance analysts, Product
- Marketing analysts, and decision-makers who consume or produce dashboards.
Pre- requisites
- Comfortable with Excel basics (filters, formulas).
- Familiarity with basic charts and descriptive statistics is helpful.
- Software installed: Power BI Desktop or Tableau Desktop/Prep (trial is fine).
- Sample datasets will be provided (sales, operations, and customer data).
Course Content
Clean → Model → Summarize
Module 1: Framing for Decisions
- Clarifying decision questions & metrics; defining “good summary” vs raw tables
- Data quality dimensions and a simple QA checklist
Module 2: Data Ingestion & Profiling
- Connecting to CSV/Excel/Sheets/Databases
- Profiling columns: data types, distributions, outliers, nulls
Module 3: Data Cleaning & Reshaping
- Power Query / Tableau Prep: remove duplicates, standardize categories, trim/split/merge
- Pivot vs Unpivot; handling multi-sheet files; appending monthly files
- Joining tables (inner/left/right); fuzzy matching basics
- Hands-on Lab: Build a repeatable cleaning pipeline
Module 4: Modeling for Analysis
- Star schema basics: facts, dimensions, surrogate keys
- Relationships, filter direction, granularity pitfalls
- Date tables & time intelligence setup
- Hands-on Lab: Create a simple model ready for KPIs
Module 5: Summaries & Business Measures
- Aggregations: SUM, AVERAGE, DISTINCTCOUNT, weighted metrics
- Power BI (DAX): CALCULATE, FILTER, IF, DIVIDE, quick measures; time intelligence (YTD, YoY, rolling 90 days)
- Tableau: Row-level vs aggregate calcs, Table Calcs, LOD (FIXED/INCLUDE/EXCLUDE) for stable KPIs
- Hands-on Lab: Build KPI cards and summary tables (Top N, variance vs target)
Visualize → Interact → Communicate
Module 6: Visualization Best Practices
- Choosing the right chart for the question
- Reducing clutter; labeling; color use for meaning; small multiples vs single complex charts
- Accessibility & consistency (fonts, spacing, contrast)
Module 7: Dashboards & Interactivity
- Layout grids, visual hierarchy, white space discipline
- Slicers/filters, drill-through, bookmarks (Power BI); actions, highlight, parameters (Tableau)
- Tooltips for context; designing “guided analysis”
- Hands-on Lab: Build an interactive executive dashboard
Module 8: Data Stories for Decisions
- Structuring a narrative (setup → insight → implication → action)
- Annotating insights; adding targets/benchmarks; scenario and what-if basics
- Hands-on Lab: Present a 3-minute insight story
Module 9: Publishing, Sharing, and Refresh
- Power BI Service / Tableau Server/Cloud overview
- Data refresh, gateway basics, versioning
- Row-Level Security (intro) and permission patterns
Module 10: Performance & Validation
- Model size tips, aggregation tables, query reduction
- Cross-checking against source totals; unit tests for measures; documenting lineage
Capstone Project (Team)
- Start from a messy, multi-file dataset → clean → model → KPIs → dashboard → 5-slide executive readout
- Instructor critique focused on decision-readiness and clarity
