Synopsis
Business users rarely ask for complex data architecture; they just want fast answers. This article explores how Ergon Energy Retail used Databricks’ Genie AI to close the gap between legacy reporting and real-time insight, allowing their teams to interrogate customer data directly through natural language.
Bridging the Reporting Bottleneck: How Ergon moved past the old style of waiting seven weeks for a simple report by giving business leaders direct access to their data.
AI as “Training Wheels”: Why Natural Language Querying (NLQ) isn’t just a gimmick, but a tool that removes the fear of breaking systems and builds user confidence.
“Veridating”: How the focus at Ergon has shifted from data literacy to business literacy, requiring users to explicitly verify and validate AI outputs.
Lessons for Data Leaders: Why executive sponsorship is non-negotiable, and how balancing AI exploration with centralised reporting turns data into a competitive edge.
Damien Lynch, General Manager Customer and Business Enablement and Huy Tran, Manager, Reporting, Delivery and Business Partners from Ergon Energy Retail joined us to discuss their journey from legacy reporting to a modern, AI-powered data platform and all the hiccups in between.
At this Breakfast Club by Ignite, the actual business user, Ross Lythall, Operations Manager at Ergon Energy Retail was in the audience and was quickly drawn into the conversation.
The Catalyst for Change
For nearly two decades, Ergon had relied on essentially the same technical product set for its reporting. When Energy Queensland mandated a corporate move to Databricks, and the business simultaneously rolled out a new core billing platform (Kraken), there was a significant opportunity for the business to re-think its product and technology plans.
While recounting this, Damien noted a universal truth about data transformations:
Business users rarely ask for "a modern data lake house with three medallion layers". What they actually want is an SQL developer sitting on the end of their desk to get them the answers that they want when they want them. The challenge for Huy, Ross and the data team was figuring out how to surface the new platform's value without bottlenecking everything behind technical analysts.
Bridging the Gap: Finding the Hidden Value
To get the business on board, the team didn’t start with complex technical jargon. Instead, they focused on what the business understood intimately: the customer. They built a foundational model, uniting account, and customer data, which had always been their richest dataset.
Through surfacing data in a model that the business recognised and understood, the team successfully exposed the platform’s potential. As Ross pointed out, this was a massive shift from the old days, where asking for a simple report would take seven weeks, and by the time you got it, you couldn’t even remember why you needed it!
The Role of AI: “Training Wheels” for Data Analysis
One of the most surprising takeaways was the impact of Databricks’ Genie AI. While initially viewed as just another sales pitch, natural language querying became the unexpected hero of Ergon’s upskilling strategy.
Huy explained that the real barrier to adoption wasn’t just access to data, but the fear of breaking something. Users were concerned that running a massive query would bring the system down.
Genie AI removed that barrier, acting as “training wheels for data analysis”. Through enabling users to ask questions in plain English – and showing them the generated SQL code alongside the answer – the tool helped Ergon team members take their first steps into data exploration, building the confidence to start to try new things.
The results are already tangible:
- Large Excel workflows that used to take four hours to calculate are now running in minutes.
- A significant reduction in the backlog of data dashboards, as team leaders no longer need these as they can directly access and interrogate data when they need it.
Making Smart People Smarter (and “Veridating”)
A fascinating philosophical theme emerged around critical thinking.
Conversational Analytics now allows all users to ask business questions of their data more naturally. Data literacy is not as much of an obstacle as it used to be, and the focus now becomes business literacy – how do people ask better questions.
On the flip side, because users are aware that AI sometimes confidently forces an answer – they have to be focussed on critically analysing the output. Ross coined the perfect term for this new user responsibility: “Veridate” (verify and validate). His team are now confidently interrogating and questioning data and its uses, understanding where it comes from and how to use it.
Executive sponsor Ayesha Razzaq, Executive General Manager of Retail at Energy Queensland (who was also in the audience) agreed, making a key point about AI: while an AI hallucinating might make the front page of the paper, we are using this at Ergon to drive a culture of questioning and interrogating facts.
The shift requires leaders to foster a culture where more questions are asked, and everything is checked, starting by teaching users to ask small, precise questions rather than asking the AI for the world right out of the gate.
Key Takeaways for Data Leaders
Executive sponsorship is non-negotiable
Having top-down support from leaders like Ayesha gave the project an anchor and helped shift a culture away from lack of data access as an excuse for inaction.
Centralise the formal reporting
Encourage broad exploration and conversational analytics with AI, whilst complementing that with structured data teams who can validate anything going to the executive level or external stakeholders.
Your data is a competitive edge
For Ergon, their business leaders had asked for better data and now they have it. ROI comes in many forms including improved employee engagement, better informed business decisions and enabling those keen to upskill by giving them access to advanced toolkits.