
Copel drives $3.3 million in projected revenue and AI-powered customer solutions
- Unlocked $3.3 million USD in projected annual revenue from new smart-meter subscription services offering energy efficiency insights, appliance fault detection, and consumption benchmarking using real-time Oracle billing data.
- Reduced receivables data visibility from days to minutes, enabling faster cash flow forecasting and delinquency risk management.
- Enabled AI-driven predictive models and retrieval-augmented generation (RAG) applications for personalized payment plans and smarter grid operations.
- Created packaged ERP, billing, and energy consumption data sets for insurance and security partners, enabling new home and life insurance offerings.
“One year ago, real-time analytics and GenAI were an afterthought. Today, they’re a revenue engine. Thanks to Fivetran, we’re improving cash flow and unlocking revenue opportunities without ever slowing down operations.”
— Ewerton Sava, Director of Engineering & IT Architecture at Copel
Companhia Paranaense de Energia (Copel) is one of Brazil’s largest utility companies, serving 5.2 million customers across the country. Known for its focus on innovation and customer service, Copel set an ambitious goal: power real-time analytics and AI-driven services to unlock new revenue streams and enhance customer experiences.
Achieving this vision required rethinking how data flowed across the business. Oracle billing databases produced enormous amounts of data each week — an immense asset but also a technical challenge. Traditional replication tools couldn’t keep pace with the scale, often requiring engineering-intensive workarounds, struggling to maintain transactional consistency, and lacking automatic failure recovery. Finance teams made cash flow decisions based on week-old snapshots, delaying collections and impacting working capital. Product teams were unable to launch AI-powered features or new revenue-generating offerings.
The team needed a solution that could replicate data at scale without disrupting mission-critical operations, while also supporting long-term flexibility for new use cases like smart meter streaming. Working with systems integrator BlueShift Brasil — who helped Copel organize its Google Cloud data lake, understand the structure of its legacy systems, and design cross-system data models — Copel evaluated several data integration tools and chose Fivetran for its winning combination of speed, reliability, and simplicity. BlueShift’s deep technical partnership ensured Copel could modernize its architecture quickly and confidently, enabling seamless ingestion across both operational and new AI-driven workloads.
Real-time data accelerates cash flow
With Fivetran, Copel quickly replicated data from its 6 Oracle billing databases to Google BigQuery in under 4 weeks, all without interrupting daily operations.
Fivetran’s log-based change data capture (CDC) replicates more than 400 GB per hour of redo logs, updating BigQuery every 5 minutes with minimal impact on source systems. Minimally invasive, high-volume data replication keeps pipelines reliable even as Oracle data volumes grow, giving Copel teams continuous access to the data they need to act quickly and confidently.
With Fivetran, Copel is able to turn information into immediate action:
- Account receivables are visible in minutes, not days, giving finance teams the confidence to forecast cash flow and act on delinquency risk before it escalates.
- Leadership has a real-time view of company performance through Power BI dashboards.
- AI-ready data feeds Vertex AI models that predict payment behavior and recommend tailored payment plans.
“With 5-minute latency and a scalable cloud platform, we can act on data quickly, improving our cash flow posture and monetizing our data to drive revenue growth without fear of blocking the operations.”
— Ewerton Sava, Director of Engineering & IT Architecture at Copel
Unlocking revenue growth and smarter customer solutions
Armed with trusted, fresh data, Copel is moving beyond traditional utility services. Product teams are designing new subscription offerings projected to generate $3.3 million USD (approximately R$18 million) in annual revenue starting in 2025.
The company is also turning its ERP, billing, weather, and energy consumption data into packaged insights that insurance and security partners use to tailor home and life insurance offerings, creating new revenue streams beyond its core utility services. A newly formed AI team leverages this same data foundation to build predictive models that forecast payment delinquency, detect energy fraud, identify equipment anomalies, and support grid-reliability planning — laying the groundwork for smarter, data-driven operations across the utility organization.
Looking ahead, Copel plans to stream smart meter IoT data through Fivetran to give customers deeper visibility into their consumption patterns, support regulated usage-based tariff models, and help customers identify opportunities to reduce their energy costs. This same real-time data foundation will also enable proactive outage detection and other innovative customer-facing services.
[CTA_MODULE]

Découvrez comment l'automatisation du mouvement de data stimule la productivité et accélère l’obtention d’informations pour votre entreprise.
Télécharger le rapport
Découvrez comment Fivetran booste l’IA et l’analytique
Télécharger le guide



.png)
.png)





