Elevating data in the race to unlock the AI opportunity

It’s an exciting time to be working with data in business. With new AI tools unlocking incredible opportunities to put data to work, there’s a renewed energy for those of us working in technology as another big change in how we work sweeps through the industry.

Of course, there’s a fine line between being excited or feeling overwhelmed by what’s happening. Reports suggest that only 4% of local organisations feel prepared to leverage AI. What’s holding us back? The primary readiness barrier is the state of existing organisational data. Not just the question of data infrastructure, which is important. Even bigger is the question of how the business culture itself approaches data management.

The good news is that we’ve never seen more interest amongst executives in getting our data in order. Between the desire to leverage AI and the news of major data breaches in recent years, there’s energy to get our data house in order supported from the very top. And what serves us well from a cybersecurity perspective – knowing our critical data elements and how we protect them – also serves us in building readiness for AI integration.

There are three key factors in elevating data maturity in an organisation. Simplify the data landscape and preventing data sprawl to build confidence in data as a source of truth. Cultivate a data-driven culture throughout the organisation. And finally, ensure there is a value led approach to our data strategies to measure the impact they deliver for the business. Together, these form the heart of data maturity and create the conditions for AI-fuelled success.

We think of data as the lifeblood of the organisation, not something that only lives in a data warehouse. It sits in every application that is being used around an organisation. It is constantly in flow and can often be duplicated and poorly controlled. A lot of the foundational work we’ve done at Telstra has been more than about tagging data or protecting it in data warehouses. It’s been about simplifying our data ecosystem, so we don’t have to work so hard to manage and protect it.

This has meant approaching data governance as more than a policy and procedure document. It’s about establishing a culture, just like cybersecurity today, where everyone understands they are responsible for data management. When everyone shares in protecting, simplifying, and managing appropriate access to data, we can begin to unlock its full potential for its responsible use within an AI-supported business.

Any work we do with partners and customers preparing for AI integration is being built upon the experience from the journey we’ve undertaken within Telstra. We set ourselves a goal to improve 100% of our key business processes using AI by FY2025. In practice, this means also rethinking every product, service, and process for the way AI will fuel how we will live and work in the future. 

Inside Telstra, early on we identified the opportunities where the adoption of AI would help achieve business process efficiency and we are currently on track to improve 100% of our key business processes by using AI by FY25. Now we believe there is emerging opportunity to use AI to help with data challenges. We’re exploring the use of Large Language Models to analyse, classify and simplify our existing data holdings. We’re also exploring the use of Generative AI to draft missing documents in our frontline knowledge base. We’re using AI to help improve our data, so it is AI-ready. 

This approach will be particularly valuable for one of our flagship uses of AI. As part of finding real challenges to solve with AI, we looked at pain points appearing in employee engagement surveys. We saw an opportunity to help frontline staff across contact centres and in our stores answer customer questions more easily. There is a lot of information that frontline teams need to access – from pricing to products to procedures - and being able to answer a customer query in the moment can be a very manual and time-consuming process. We found generative AI a perfect fit for this task, and an excellent opportunity to optimise knowledge management to surface the right information quickly and intuitively. We call this tool Ask Telstra and it has been a great boost to employee experience for frontline team members involved in its trial in late 2024 and subsequent roll out this year.

We’ve also applied AI to help team members quickly get up to speed with what a customer has recently been in contact with us for. One Sentence Summary transforms recent customer notes, interactions, and transactions into a concise summary of a customer’s recent history and status, reducing the need for customers to repeat information. We trialled this capability in Canberra and the team loved it so much we’ve now launched it for all our contact centre agents.

This process also established the importance of keeping our knowledge management articles up to date and accurate at all times. And more broadly, that all data needs to be treated in ways that give everyone confidence it can be trusted as a reference for whatever tool needs it in future. The technical process of making these documents available to AI must be supported by the cultural underpinning that data must be responsibly maintained.

Elevating the value we place on data also means increasing all efforts around how we protect it. This includes not only security concerns but also the amount of data we purge altogether as we aim to minimise data sets and truly simplify our data holdings. Our thinking around the data we collect, and our use of AI is not only a question of staying within our legal obligations but also of meeting the expectations of the Australian community. 

Our customers are telling us how important AI is becoming in supporting sustainable and safe practice, particularly in the mining sector. AI is supporting vision and sensor systems in mines to monitor data for regulatory compliance reporting. We are using similar AI systems to monitor for possible degradation in our networks to take proactive action before they cause any impact on our customers. These kinds of large-scale AI monitoring tools give organisations the ability to move from sample-based checks for safety or reliability and begin to analyse every instance across millions of data points.

We are not alone on this journey towards becoming an AI fuelled organisation. Our experts are constantly working with technical partners across a range of data and AI toolsets to find the best -fit solution for any given task. We are in the very early days of AI. We anticipate rapid evolution of technologies and opportunities. To support us in the safe and scaled adoption of AI we’re building industry-leading expertise in the safe and responsible use of AI.  This also helps us to bring solutions to our customers to help them find the best path for their own needs, whether in customer service, primary industries, or knowledge management.

Achieving AI readiness can feel daunting. But we recognise there is opportunity to leverage AI to help us overcome our previous data challenges. With so much to be gained, it’s the perfect time to move from feeling overwhelmed to feeling excited by what’s possible.