We caught up with Mark Bonney, VP IT, Change and Data at SmartestEnergy Australia, a year since we last met, to see how his program of ‘Prototype to Production’ was working.
A little about the journey so far
Mark describes walking into the role at SmartestEnergy Australia as ‘walking onto fertile ground’. Mark started two years ago at SmartestEnergy Australia when there were only 50 people. And he started with a leadership team that recognised the role that data would play in the business and from day one, he had an important ‘seat at the table’.
While the SmartestEnergy business in Australia was still in its start-up phase, they had the benefit of a 20-year-old organisation, some 500 strong, based in the UK to lean on. Learning from mistakes of the past, SmartestEnergy Australia had decided to invest in a pure data role, reporting straight into the CEO. Fertile ground.
The SmartestEnergy Australia business was 100% cloud-based systems, had two developers in the IT team and seven citizen developers in the business.
The citizen developers were busy building an integrated trading system to meet their immediate needs, including forecasting, price curves, trying to pull down meter data and trying to create trading positions and reports. It was a largely uncontrolled SQL server database environment, and the analysts were ‘knocking stuff’ together. It was what Mark calls prototyping in a simple, uncontrolled manner.
At the same time, SmartestEnergy Australia’s IT team only focused on the retail side of the business. They were really trying to settle Salesforce in at the fundamental level.
As the new leader, he was less worried about the retail side, he thought, “The trading, risk and finance areas are heading to ‘I can’t reconcile anything’ territory.”
This was a tale that we have all seen many times before, in businesses across all industries. Mark and his team had to understand why the citizen developers were building what they were building. They were super capable, knew the business really well, had the trust of their leaders, and were creating tremendous value. But the IT team wasn’t even at the table.
It's always about the people
Enter Ignite
Ignite started by working across the business to really understand the business personas and the different ways people were working.
Then Mark had to make a technology choice to support the operating model. The global business had defined Synapse as the platform of choice. The citizen developers were comfortable writing SQL, however, traditional SQL databases don’t scale processing time series data. Synapse was a great opportunity to be unconstrained from a storage and processing perspective.
So, how do you get anyone to move across to a Synapse environment, which, although technically superior, is more complex and required learning new skills, on top of their day job?
Mark asked Ignite’s Kris Hunter, to come in. He talked to everyone about a collaborative way of working alongside IT, and creating a fit-for-purpose technology platform.
The operating model included an explore zone where SmartestEnergy Australia would enable their citizen developers to do what they like, with a light level of governance over the activities. In this environment, IT was accountable for providing data as a service.
Data as a Service
IT could then focus on providing reliable data sources that had been reconciled to the markets and could be trusted. In the explore zone, people could start to build things, in a new technology with IT supporting the environment. They put people & operating model first, technology choice second.
Then Mark realised that key trading analysts were worrying about whether their data had loaded each morning. He found it I easy to convince them and their bosses that he could take this work into the IT team and release them from these production tasks.
Once they established this base for the citizen developers – that they could trust what the IT team made for them, Mark and his team started to create another service called prototype to production. IT will pick up anything that is written, stabilise and productionise it; if the business case is signed off.
"We started by picking up and migrating forward curves from the explore zone into production – making robust, fast, reliable, supportable solutions. We developed a simplified change process to support updates. If they then want to craft version two, they can do that in the explore zone and hand it over when its ready. We had cases where reports that had taken the citizen developer 4 hours to run, turned to five minutes once it had moved to production, and can be run many times in a day at limited incremental cost."
So, the IT team was able to get a license, prove their value through this exercise, getting everyone onto one platform in one shared environment and investing in the data.
Why it works
Mark’s team wrote business cases that were mostly benefit positive, establishing the real business benefits, that improved the quality of forecasts and revenue assurance. These were then approved and prioritised by the Executive Team. And then every month, reported on and prioritisation confirmed.
And the executives liked the language they were using – assuring data quality and accuracy. They worked hard to help them to recognise the value of committing to and supporting data projects.
And they had a good groundswell of underlying support from the citizen developers, who had found the support really beneficial to them and their work. They were able to make better business decisions, free up time and focus on analysis.
Mark’s team had no problem getting people to put their hands up for projects, because they could see the value that we are delivering.
Importantly, they don’t touch the prototypes until they have reached a point of stability and maturity.
And now, there is a strong sense of collaboration across the citizen developers and the IT teams. A real sense that the process works and they can move fast and deliver quality
There are now six people in IT – three developers and three analysts serving ten citizen developers within the business. They are now building tools on top of the data. They are even starting to think about how they could take these tools to other parts of their international business – the UK and US.
"I think my job is to make everyone in the organisation a developer if they want to be – the business knows what it needs to answer and IT can make those answers accurate, fast and reliable. Every graduate who walks in our doors can code – so we are already tech savvy, data is really – everyone’s business."
