What challenges did Copel face before AI implementation?

Copel needed timely, accurate insights from backend SAP ERP data. Earlier workflows relied on manual processes, RPA, and Excel exports, which were time-consuming, inconsistent, and could take up to a day.

What was the impact of the AI solution?

Query processing time reduced from one day to near real-time/seconds. Manual and time-consuming steps were eliminated. The solution improved productivity and decision-making and supported data democratization. The article also states that significant cost savings were realized.

Source

GCPEvidence: Low40/100

Copel GenAI virtual agent for querying SAP ERP with Gemini, Vertex AI, BigQuery and Cortex

Use case typeAI platformUpdated Jun 13, 2026

Copel, a leading Brazilian energy company, transformed its data access strategy with Google Cloud generative AI technologies and large language models. The company built a virtual agent that connects to its on-premises SAP ERP system and lets employees ask natural-language questions to retrieve timely insights from enterprise data. The solution replaced manual processes involving RPA tools and Excel exports, improving speed, consistency, productivity, and decision-making.

Organization: Copel
Industry: Energy & Utilities
Location: Brazil
Published: May 2026

Reported outcomes

Strategic outcomes

New product / capabilityBuilt a natural-language ERP query agentBetter decisions & insightImproved productivity and decision-makingCost efficiencyEliminated manual and time-consuming steps

Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks

Customer: Copel
Provider: GCP
Maturity: Unknown
Linked source: Google Cloud Customer Story

No explicit deployment-stage evidence found.

Customer identity supportedSource describes one deploymentMaturity evidence evaluated

Primary read

Use case focus

Showing 3 of 4

1Generative AI
2Enterprise Search
3Workflow Automation

Copel needed timely, accurate insights from backend SAP ERP data.
Earlier workflows relied on manual processes, RPA, and Excel exports, which were time-consuming, inconsistent, and could take up to a day.

Copel developed a virtual agent powered by Gemini Pro 1.5 and Vertex AI.
The agent connects to on-premises SAP ERP and uses BigQuery, Dataplex, and Cortex to support enterprise data access and answer employee questions in natural language.
The system delivers answers in seconds instead of days.

Technologies

Gemini Vertex AI BigQuery Dataplex Cortex

Query processing time reduced from one day to near real-time/seconds.
Manual and time-consuming steps were eliminated.
The solution improved productivity and decision-making and supported data democratization.
The article also states that significant cost savings were realized.

Sources & evidence1

Evidence: Low40/100Evidence strength