NovaPay Dashboard: A Fintech Analytics Case Study

Introduction

Dashboards are only valuable when they help people make better decisions. When I started building the NovaPay dashboard, my objective wasn't simply to visualize data. I wanted to design a decision-support tool that answered the questions a fintech leadership team would ask every day: Where is revenue coming from? Which products are driving growth? Where are the biggest risks? Which channels deserve more investment? And what actions should the business take next?

NovaPay is a fictional Nigerian fintech offering a portfolio of consumer and SME financial products, including Personal Loans, Business Loans, Buy Now Pay Later (BNPL), Wallet & Transfers, Bill Payments, Micro-Insurance, and Savings. The dashboard covers 24 months of business activity from January 2023 to December 2024, bringing together data across six business functions. It analyses more than 53,000 customers, 875,000 transactions, 159,000 loans, and a ₦207.7 billion sales pipeline, providing executives with a single view of the company's performance.

More than building an attractive dashboard, this project challenged me to combine data modelling, DAX, user experience, and business storytelling into a solution that helps decision-makers move from information to action.

Design Objectives

Before building a single visual, I defined a simple design goal:

Design a dashboard that executives could use to understand performance and make informed decisions.

Many dashboards overwhelm users with colourful charts, inconsistent formatting, crowded layouts, and visuals that compete for attention. I wanted the opposite: a clean interface where every element had a purpose. The design uses a dark fintech-inspired theme built around teal accents (#5FDFDC and #32707E) on a charcoal background (#111827). White typography creates strong contrast for KPIs, while teal highlights guide attention toward important metrics without overwhelming the page.

Consistency became a guiding principle throughout the project. Every page follows the same layout: headline KPI cards at the top, a contextual insight banner summarising the most important takeaway, slicers for exploration, and supporting visualisations organised to tell a coherent story.

Rather than asking users to interpret every chart individually, the dashboard provides immediate context through these executive insight banners. Each page begins with a concise summary of the most important business finding before users explore the supporting visuals. This allows executives to understand the key message within seconds while giving analysts the flexibility to investigate further. The result is a dashboard designed for decision-making rather than data exploration alone.

Dashboard Structure

The dashboard is organised into six interconnected pages, each representing a key business function while contributing to a single executive narrative.

The Home page serves as an executive overview, bringing together headline KPIs and key insights from every department. It provides leadership with a quick snapshot of overall business performance and acts as the primary navigation hub.

The Finance page focuses on revenue performance, transaction activity, fee income, market share, and geographic distribution. It highlights where revenue is generated and identifies opportunities for diversification.

---

The Loan page provides visibility into portfolio performance, repayment health, outstanding balances, default rates, and loan concentration across products and locations, enabling executives to monitor both growth and credit risk.

---

The Marketing page evaluates acquisition performance across channels, comparing customer acquisition costs, campaign effectiveness, conversion rates, and marketing spend to identify the most efficient growth opportunities.

---

The Retention page measures customer health through churn, customer satisfaction, Net Promoter Score (NPS), support tickets, and customer status distribution, helping identify potential risks to long-term customer value.

Finally, the Sales page tracks pipeline performance, deal progression, win rates, sales cycles, and sales representative performance, providing visibility into commercial execution.

Together, these pages create a complete picture of NovaPay's business performance while allowing users to explore each functional area in greater detail through contextual slicers and interactive filtering.

Data Model and DAX

Behind every effective dashboard is a well-structured data model. NovaPay brings together six core datasets covering customers, transactions, loans, marketing, retention, and market performance. These tables are connected through customer and date relationships, supported by a dedicated calendar table that enables consistent time-based analysis across every page.

When developing measures, I deliberately favoured simplicity and readability over unnecessary complexity. Measures such as Conversion Rate, Churn Rate, Customer Acquisition Cost, and Success Rate rely on straightforward DAX calculations that are easy to understand, validate, and maintain. In my experience, maintainable models are usually more valuable than overly complex formulas.

One interesting challenge involved calculating NovaPay's market share value. The calculation combined a predefined market share percentage with the estimated total addressable market, producing figures in the tens of trillions of naira. Ensuring those values displayed correctly without scientific notation required additional formatting, reinforcing an important lesson: presentation is just as important as calculation when building dashboards for executives.

Key Business Insights

One of the most rewarding parts of building the dashboard was analysing the completed model and identifying the business story hidden within the data. The analysis revealed several important insights.

Revenue concentration presents a strategic risk. Personal Loans and Business Loans generate approximately 55% of total fee income, while Lagos alone contributes roughly 40% of revenue. Although these products currently drive growth, this level of concentration exposes the business to regulatory changes, increased competition, or regional disruptions. Diversifying both product adoption and geographic reach should become a strategic priority.

The loan portfolio is heavily concentrated in Business Loans. Approximately 86% of outstanding balances come from a single product category. While the overall default rate remains healthy, such concentration increases exposure to sector-specific credit risk and highlights the importance of ongoing stress testing and portfolio diversification.

Social Media delivers the strongest marketing performance. It acquires the highest number of customers while maintaining the lowest customer acquisition cost. By comparison, Referral receives the highest marketing spend but generates fewer customers at a higher acquisition cost, suggesting a clear opportunity to optimise marketing budget allocation.

Customer churn appears to be a business-wide challenge rather than a segment-specific issue. Churn remains relatively consistent across customer groups, suggesting that retention challenges are more likely linked to product experience or service quality than to any single demographic segment.

Sales performance is highly uneven across representatives. A small group of top-performing sales representatives contributes a disproportionately large share of total deal value, highlighting opportunities to improve coaching, territory allocation, and sales enablement across the wider team.

Design Decisions

Several design decisions had a significant impact on the final dashboard. One of the most important was separating Finance and Loan performance into independent pages. Combining both subjects on a single page created unnecessary visual density and reduced readability. Splitting them allowed each page to tell a clearer story while giving individual charts more space.

Another decision was removing duplicate visualisations. During early iterations, multiple charts communicated essentially the same information from different angles. Rather than increasing visual complexity, I replaced redundant charts with visuals that introduced genuinely new insights, such as revenue distribution by city.

I also adopted semantic colour coding throughout the dashboard. Teal represents primary metrics, while red is reserved for alerts such as default rates or potential business risks. This helps users interpret information more quickly without needing to read every number in detail.

Finally, I incorporated interactive tooltips into several visualisations. Instead of simply displaying values on hover, these tooltips provide additional context such as monthly trends, top-performing segments, default rates, and geographic performance, allowing users to explore the data without leaving the current page.

Future Enhancements

If NovaPay were a production system rather than a portfolio project, several enhancements would further increase its value. Potential additions include a predictive churn model to identify customers at risk of leaving, Customer Lifetime Value (CLV) calculations integrated with acquisition costs, scenario modelling for revenue and credit risk planning, and anomaly detection to identify unusual transaction patterns or potential fraud in near real time. These additions would shift the dashboard beyond descriptive analytics toward predictive and prescriptive decision support.

Conclusion

This project was about more than building a visually appealing dashboard. It demonstrates how I approach analytics projects, from designing an intuitive user experience and building a maintainable data model to writing efficient DAX and translating business metrics into actionable insights. Every page, KPI, visual, and interaction was designed with a clear purpose: to help decision-makers understand performance, identify opportunities, and make better business decisions. For me, that's what effective analytics should do. It shouldn't just present data; it should make the next decision easier.