QuickSight Dashboard & Analysis | Overview.

QuickSight  Dashboard & Analysis - Overview. 

Scope:

• Intro,
• QuickSight Overview,
• QuickSight Terminology,
• Analysis vs Dashboard,
• Workflow,
• QuickSight Architecture Sample,
• Visualization Sample,
• Best Practices,
• QuickSight Pro Tips.

1. QuickSight Overview

Amazon QuickSight is a cloud-based BI service from AWS that allows twtech to:

• Connect to multiple data sources (S3, RDS, Redshift, Athena, etc.)
• Analyze data with interactive visualizations
• Share insights through dashboards
• Use ML insights like anomaly detection and forecasting

Key features:

• SPICE (Super-fast, Parallel, In-memory Calculation Engine): Accelerates analytics.
• Row-level security: Fine-grained data access.
• Embedded analytics: Integrate dashboards into apps.
• ML Insights: Auto-narratives, anomaly detection, forecasting.

2. QuickSight Terminology

Term	Description
Analysis		The “workspace” where you create visuals and explore data. Can have multiple sheets and visuals.
Dashboard		A published version of an Analysis, intended for consumption by end-users. Interactivity is limited compared to Analysis.
Dataset		Connection to a data source, which can be direct query or SPICE-based.
Data Source		Where the data lives (S3, Athena, Redshift, RDS, etc.).
SPICE		In-memory engine that caches datasets for fast queries.

3. Analysis vs Dashboard

Feature	Analysis		Dashboard
Purpose		Build and explore visuals	Share insights with end-users
Editability		Fully editable	Read-only for users (except filters/interactions)
Sheets		Multiple sheets possible	Same sheets from analysis, but no new sheet creation
Interactivity		Full (filters, drill-downs, calculated fields)	Limited (filters, drill-downs if enabled)
Audience		Analysts, power users	Decision-makers, executives

Workflow:

1. Connect to dataset → 2. Build analysis → 3. Publish as dashboard → 4. Share with users

4. QuickSight Architecture Sample

Let’s take a practical scenario:

Goal: Build a Sales Dashboard using S3 + Athena + SPICE with row-level security.

Step-by-step Architecture

1. Data Storage (S3)
• Store raw CSV or Parquet sales data in S3 buckets.
2. Query Layer (Athena)
• Use Athena to query S3 data using standard SQL.
• Example: aggregate monthly sales by region/product.
3. Dataset in QuickSight
• Create a dataset pointing to Athena queries.
• Import to SPICE for fast performance or use Direct Query for real-time data.
4. Row-Level Security (RLS)
• Define user-to-data mappings (e.g., regional sales managers can only see their region).
• Attach RLS rules to the dataset.
5. Analysis
• Build visuals: bar charts, KPIs, heatmaps.
• Add calculated fields: YoY growth, conversion rate.
6. Dashboard
• Publish analysis as dashboard.
• Enable interactive filters: region, product category, time period.
• Share with stakeholders via email or embedded app.

Visualization Sample:

• KPIs: Total Sales, Total Orders
• Charts: Sales Trend, Product Performance, Regional Heatmap
• Filters: Date, Region, Product Category

5. Best Practices

1. Use SPICE when possible for performance and concurrency.
2. Use Direct Query for live data that changes frequently.
3. Implement RLS to enforce data governance.
4. Optimize dataset design: pre-aggregate data in Athena/Redshift if needed.
5. Minimize visuals per dashboard: 10-15 max per sheet for clarity.
6. Use themes & consistent color palettes for usability.
7. Test interactivity: Filters, drill-downs, and control actions before publishing.

6. QuickSight Pro Tips

• Calculated fields: Can create complex KPIs without changing source data.
• Dynamic Titles: Update dashboard titles based on selected filters.
• Parameters: Make dashboards interactive (e.g., user selects year/month).
• Embedding: QuickSight dashboards can be embedded into apps for end-users.