Start Where You Are

A dashboard is not just a collection of charts; it is a single-screen display of the most critical information required to do a job. In business, a well-designed dashboard helps you view data at a glance, collaborate with stakeholders, and make strategic decisions based on accessible evidence.

We distinguish between three main types of dashboards:

• Operational: For day-to-day tracking and short-term decisions (e.g., "Are we out of stock of Star Wars?").
• Analytical: For deep dives into the "why" and "how" (e.g., "Which genres correlate with high late fees?").
• Strategic: For executives monitoring high-level KPIs and progress toward long-term goals (e.g., "Is our monthly revenue per customer growing?").

Our focus in this lab is Strategic. You are not just reporting numbers; you are telling a narrative about whether the store's strategy is succeeding.

Dashboard Design Principles

To move beyond a "data dump" and into effective storytelling, apply these core principles:

• The Prime Real Estate Rule: Users scan a page from top-left to bottom-right (the F and Z patterns). Place your most critical headline KPI—Monthly Revenue Per Customer—in the top-left corner.
• Progressive Disclosure: Start with big, bold numbers (headline KPIs) to show what is happening, then follow with charts that explain why (e.g., a breakdown by film category).
• Volumetric Consistency: Your dashboard must reflect the same "Doctrine of Volumetric Integrity" we established in Lab C. If your row counts changed during ETL, ensure your dashboard totals (e.g., total customers) align with those known counts to maintain trust in the data.
• Minimize "Non-Data Ink": Avoid unnecessary decorations or artsy charts that don't add meaning. Choose the simplest chart (bar, line, or KPI card) that clearly communicates the message.

<aside> 💡

Inset: It’s Never Really "Done"

A high-quality dashboard is not a static monument to data; it is a collaborative, iterative analytic product. If a dashboard is truly effective, it doesn't just provide answers—it creates better, more sophisticated questions.

• The Cycle of Inquiry: When a manager sees that "Monthly Revenue Per Customer" has spiked, their next question isn't "What is the number?" but "Which specific customer segment drove that change?" This often requires an additional drill-down dashboard or the acquisition of new data sources to explore the "why" behind the "what".
• The Co-Creation Mandate: A dashboard should never be built in a vacuum. It must be co-created with the end-user. As the user begins to rely on the tool, you will notice a "streamlining" effect: certain numbers that seemed vital during the design phase will fade from view, while a "favorite metric" emerges as the daily pulse of the operation.
• The Designer’s Burden: While the technical capabilities of a tool (like Domo or Power BI) are important, the majority of a dashboard’s value lies in the designer and maintainer. Your job is to curate the experience, pruning away the noise and ensuring the product remains relevant as the business strategy shifts.

Remember: An analytic product is in a state of constant refinement as long as the business process it supports continues to be performed.

</aside>

Reflection Questions

1. Audience Persona: Who is the specific "owner" of this dashboard? Is it the store manager or the corporate CFO? How does their role change the level of detail you show?
2. Lagging vs. Leading Indicators: Monthly revenue is a lagging indicator (it shows what already happened). What is one leading indicator (a metric that predicts future revenue) you could include to give management an early warning?
3. The "Why" Behind the Nulls: In Lab C, we used Left Joins to include customers who haven't rented yet. How should these "inactive" customers be visualized on your dashboard? Do they represent a problem or an opportunity?
4. Strategic Alignment: If your dashboard shows revenue is increasing but customer satisfaction (NPS) is dropping, is the strategic initiative truly successful in the long term?