Video summary

Tableau Banking Data Analytics End to End Project | Tableau Realtime Project Explain Hindi #analysis

Main summary

Key takeaways

Technology

Summary of the Video (Tableau Banking Data Analytics End-to-End Project | Hindi)

The video walks through an end-to-end Tableau banking analytics project aimed at building interview-ready skills. It explains both the project lifecycle (from requirements to delivery) and a hands-on Tableau walkthrough for creating multiple dashboard views, KPIs, filters, and exporting/testing/documentation.

1) End-to-end Project Workflow (How the Project Is Delivered)

The speaker emphasizes this structure is useful for interviews and real-world BI delivery:

  1. Requirement Gathering

    • Identify who sends requirements and where they come from (clients/stakeholders).
    • Clarify business goals (e.g., profit, revenue, sales) and what dashboards should show.
    • A Business Requirement Document (BRD) is created by the Business Analyst.

  2. Data Acquisition

    • Determine where the data comes from (Excel, database/server/SQL sources, etc.).

  3. Data Cleaning & Transformation

    • Clean and validate fields (e.g., check mismatching/invalid values using filters).
    • Prepare keys and dimensions needed for analysis, such as:
      • unique IDs
      • stakeholder fields
      • city
      • brokers
      • premium-related and profit/loss-related fields
    • Apply transformations if required (the video focuses mainly on cleaning/validation).

  4. Data Analysis & Dashboard Development (Core Tableau Work)

    • Build multiple sheets/views first, then assemble them into a dashboard.
    • Cover Tableau concepts like measures/dimensions, sorting, formatting, color palettes, and more.

  5. Documentation

    • Treat the dashboard like a product that needs a user guide, similar to a phone/laptop manual.
    • Includes how to use filters, views, and export.

  6. Testing

    • Unit Testing (UT): verify numeric correctness internally (developer + colleagues/peer).
    • Acceptance Testing (AT / “stakeholder testing”): stakeholders test the published dashboard; feedback leads to fixes.
    • Maintain a test case document with inputs and expected vs. actual results.

  7. Publishing / Packaging

    • Publish the dashboard and provide the user guide + supporting documentation.

2) Tableau Tutorial: Building the Dashboard Views Step-by-Step

A) Connecting Data & Understanding Schema

  • Use Tableau Public (free) and connect an Excel dataset (e.g., “Finance Records”).
  • Introduce the Data Dictionary, which shows:
    • field names
    • data types
    • source tables

B) View 1: Compare Total Value (Sorted Bar-Style)

  • Create a view titled along the lines of “Comparison of Total Value …”
  • Techniques used:
    • Sorting ascending/descending to highlight smallest/largest values
    • Title formatting (font, shading color), hiding unnecessary headers
    • Converting number units (e.g., millions/billions)
    • Applying a consistent company-style color palette (including mention of hex colors)

C) View 2: Premium Over Time (Trend Chart)

  • Build a time-based visualization:
    • x-axis: date
    • y-axis: premium-related measure
  • Add categorical coloring for:
    • renewal vs single premium (via count/measure-based coloring approach)
  • Formatting:
    • units, fonts, hiding grid lines and extra labels
    • title alignment and grid line removal

D) View 3: Sharing Partners Distribution (Donut Chart Technique)

  • Goal: show distribution of Sharing Partners among Stakeholders
  • Donut chart workaround (when direct donut option isn’t available):
    • calculate a “0” measure field trick
    • duplicate it to enable donut using dual-axis and mark merging
  • Additional steps:
    • format angles into percentage (100%)
    • control decimal places (single vs double)
    • hide zeros/axes where needed
    • customize angle labels and colors

E) View 4: Relationship Between Cost Ratio & Profitability

  • Use a side-by-side bar/dual-value style approach
  • Include measures such as cost ratio and profitability
  • Formatting includes:
    • hide headers
    • number formatting to billions
    • label font and size adjustments
    • remove grid lines

F) View 5: Profitability by City and Policy Tenure (Tenure Buckets)

  • Use multiple fields:
    • Profitability
    • City
    • Policy Tenure (tenure ranges like 1–3, 3–5, etc.)
  • Techniques:
    • arrange dimensions on axes (city vs tenure)
    • color represents magnitude (light/low → dark/high)
    • labels show count-based measure
  • Interpretation focus:
    • how many policy tenures fall into each range
    • how color intensity corresponds to profitability counts

G) View 6: Asset Under Management Distribution by Stakeholders (Bar Chart)

  • Use fields like:
    • Asset Under Management
    • stakeholder role/category (e.g., equity/debt)
  • Create a bar/vertical chart and apply:
    • color palette
    • label visibility
    • hide extra headers

3) KPI Tiles: Building Top-Level Metrics

The video defines KPIs as dashboard components that stakeholders can understand quickly, typically tied to:

  • total premium collected
  • average stakeholder age
  • total value created (e.g., equity logic)
  • max value created by insurer
  • and more.

Key Tableau techniques used:

  • Place KPI values in a large formatted text / measure
  • Convert display units to millions/billions and format decimals
  • Use calculated fields for custom KPIs, such as:
    • “Value Created” conditioned on Asset Under Management = Equity
    • another KPI using MAX for a metric like “Identification … max”
  • Ensure KPI accuracy by cross-checking with underlying data.

4) Filters: Interactive Slicing of the Dashboard

The video creates filters based on dataset columns, including:

  • City
  • Stakeholder
  • Gender
  • Date range / month-year
  • Premium (range/value)
  • optionally other important fields related to profitability or net worth.

UI detail:

  • Use Show Filter and adjust filter style:
    • dropdown vs multiple value dropdown
  • Rename filter titles for readability (manual edits).

5) Dashboard Building: Containers, Layout, and Styling

A major part of the video focuses on Tableau dashboard engineering.

Containers (Critical Concept)

  • Use Horizontal and Vertical containers to lock layout behavior.
  • Benefits:
    • keeps sheets stable during resizing and dashboard use
    • helps keep KPI tiles/views aligned (including “distribute content evenly”)

Dashboard Assembly Approach

  • Create a main dashboard with nested containers:
    • top area: title + logo image + padding
    • left side: vertical filter column
    • main area: KPI tiles (top) and charts (bottom/side)
  • Add logo/image using fit and alignment controls
  • Include a download/PDF export button

Presentation Mode & Export

  • Switch to Presentation Mode and verify:
    • dashboard renders properly
    • filters are visible by name
    • PDF export works correctly

6) Testing Strategy (UT/AT) + Test Case Template Idea

The video highlights validation steps such as:

  • After calculations and formatting, confirm numbers match the source
    • compare dashboard totals to Excel/underlying data
  • Test filters:
    • manually select/unselect values and ensure visuals update correctly
  • Maintain a test document/test case structure:
    • test case ID
    • description
    • input data/columns used
    • expected result
    • actual result
    • pass/fail comment
    • debugging notes (if failing)

7) User Guide Creation (Post-Development Documentation)

The final section explains creating a user guide that includes structured metadata such as:

  • dashboard name
  • author/creator
  • created date
  • version
  • purpose
  • requirements summary (charts/KPIs used, filters)
  • steps to download/export (PDF)
  • screenshots (copied via copy/paste)
  • contact/support notes

Main Speakers / Sources

  • Main speaker: Single instructor/host (speaks in Hindi; no specific name provided in the subtitles)
  • Tools mentioned: Tableau Public, Tableau BI/dashboard concepts
  • Dataset/source context: “Finance Records” (Excel) and a bank-like stakeholder example dataset

Original video