Skill Guide

Dashboard design and data visualization (Tableau, Looker, or custom Streamlit apps)

The systematic process of transforming raw data into interactive, visual interfaces (dashboards) using specialized tools or custom code to enable rapid business intelligence, monitoring, and data-driven decision-making.

This skill directly bridges the gap between complex data infrastructure and actionable business strategy, accelerating decision cycles and making data accessible to non-technical stakeholders. It translates analytical investment into measurable business outcomes like increased operational efficiency, revenue optimization, and proactive risk management.

1 Careers

1 Categories

8.7 Avg Demand

35% Avg AI Risk

How to Learn Dashboard design and data visualization (Tableau, Looker, or custom Streamlit apps)

1. Data Fundamentals: Understand basic data types (dimensions vs. measures), data cleaning, and SQL for data extraction. 2. Visualization Grammar: Learn core chart types (bar, line, scatter, heatmap) and their proper use cases based on data relationships. 3. Tool Basics: Master the foundational interface of one tool (e.g., Tableau: connecting data, dragging fields to shelves, creating basic sheets).

1. Storytelling with Data: Move beyond single charts to building cohesive dashboard narratives with clear KPIs, filters, and actions. Practice structuring dashboards for specific user personas (executive vs. analyst). 2. Advanced Calculations & Data Modeling: Implement level-of-detail (LOD) expressions in Tableau, create custom SQL views in Looker, or use Pandas for data transformation in Streamlit. 3. Performance & Best Practices: Learn to optimize query performance, manage extract sizes, and avoid common pitfalls like cluttered visuals or misleading dual axes.

1. System & Architecture Design: Design scalable, governed dashboard ecosystems across an organization. Implement row-level security, data source certification, and version control for dashboard-as-code. 2. Advanced Interactivity & Custom Development: Build complex parameter-driven workflows, write custom Python/R scripts for advanced analytics within dashboards, or develop full-stack custom applications with Streamlit. 3. Strategic Alignment: Work directly with leadership to define KPIs, translate strategic objectives into measurable data products, and establish center-of-excellence practices for visualization standards.

Practice Projects

Beginner

Project

E-Commerce Sales Performance Dashboard

Scenario

You have a dataset of monthly sales transactions including product category, region, revenue, and units sold. Build a dashboard to answer: What are our top-performing products? Which regions are underperforming?

How to Execute

1. Connect to the dataset (CSV or Google Sheets) in Tableau/Public or Streamlit. 2. Create calculated fields for profit margin and year-over-year growth. 3. Build three core views: a time-series line chart for revenue trend, a bar chart for sales by category, and a map view by region. 4. Combine these into a single dashboard with a simple filter for product category.

Intermediate

Project

Marketing Campaign Attribution & Funnel Analysis

Scenario

A marketing team needs to understand the customer journey from ad click to conversion across multiple channels (Email, Social, Paid Search). They need to see drop-off points and calculate true cost-per-acquisition.

How to Execute

1. Model the data from multiple sources (Google Analytics, CRM, ad platforms) into a unified view with consistent IDs. 2. In Tableau, use Level-of-Detail expressions to calculate first-touch vs. last-touch attribution. 3. Build a funnel chart showing progression through key stages (Impression -> Click -> Lead -> MQL -> SQL -> Deal). 4. Implement interactive filters for date range, campaign name, and channel, ensuring cross-filtering works correctly across all sheets.

Advanced

Project

Real-Time Operations Monitoring & Alerting System

Scenario

An ops team for a SaaS platform needs a live dashboard to monitor system health (API latency, error rates), user activity, and key business metrics (active users, transactions/min). They require automatic alerts for threshold breaches.

How to Execute

1. Architect a data pipeline: Stream application logs/metrics to a real-time database (e.g., TimescaleDB, BigQuery). 2. Build a custom Streamlit application: use caching (`@st.cache_data`) for performance, plot real-time time-series with Plotly, and structure the UI with tabs for different service modules. 3. Implement server-side callbacks to push data updates. 4. Write backend logic to check thresholds and trigger alerts via Slack/email integration, embedding alert status directly into the dashboard.

Tools & Frameworks

Enterprise BI Platforms

Tableau Desktop/OnlineLooker (LookML)Microsoft Power BI

Primary tools for governed, scalable dashboarding in business environments. Choose Tableau for unparalleled visual exploration and aesthetics, Looker for its powerful semantic modeling layer (LookML) and data governance, and Power BI for deep integration with the Microsoft stack (Azure, Excel, Teams).

Custom Application Frameworks

StreamlitPlotly DashPanel (HoloViz)

Used for building highly custom, interactive data applications when off-the-shelf BI tools are insufficient. Streamlit is ideal for rapid prototyping and Python-centric data science teams. Dash offers more fine-grained control over layout and callbacks for complex apps. Panel is a powerful alternative in the PyData ecosystem.

Design & Prototyping

FigmaAdobe XDSketch

Essential for designing the wireframe, layout, and visual hierarchy of a dashboard before any data work begins, ensuring user-centric design and alignment with stakeholders.

Foundational Libraries (for custom apps)

Pandas (data wrangling)Plotly / Altair / Matplotlib (viz)SQLAlchemy (database connection)

The core building blocks for any custom data application. Mastery of Pandas is non-negotiable for data preparation. Plotly provides interactive, web-native charts; Altair offers a declarative, concise API based on Vega-Lite; SQLAlchemy is the standard for connecting to virtually any database.

Interview Questions

Answer Strategy

The answer must demonstrate a structured, user-centric discovery process, not just tool expertise. Outline a framework: 1) Conduct stakeholder interviews to define primary decisions and KPIs. 2) Perform a data audit to assess availability and quality. 3) Create low-fidelity wireframes with Figma or paper to align on layout and core visuals. 4) Build a minimal viable dashboard (MVP) with core filters and iterate based on user feedback in weekly cycles.

Answer Strategy

Tests problem-solving and empathy. The candidate should move beyond technical fixes to consider user adoption factors. The strategy should involve: 1) Direct user observation (shadowing) to see how they work. 2) Analyzing usage analytics (if available). 3) Assessing the dashboard's 'time-to-insight' and cognitive load. 4) Common issues: poor performance, lack of narrative, no clear call-to-action, or wrong persona targeting.