The Evergreen Factors that Dictate AI’s Success

In the digital era, AI is a new and powerful accelerant to drive change and productivity – so that progressively more can be achieved with the same resource level. For many Australian businesses, artificial intelligence is how they will launch the next wave of productive growth.

As Atturra’s CEO Stephen Kowal said recently, “The question now is not if AI will reshape our businesses — it’s when, and how quickly and effectively we adapt. We’re continuously engaged by clients to improve almost every element of their businesses, from the simplest process to redesigning entire business models. Our only consideration as business leaders navigating this inevitable transition should be managing the change and the new future it creates.”

Buoyant predictions of Australia’s AI opportunity have not abated. Figures from one recent study by five of the country’s tech industry associations suggest AI could deliver nearly $142 billion in value to the Australian economy by 2030 by capitalising on three key areas – adoption, domestic capability, and establishing itself as a regional hub for the technology. Effective AI adoption in the finance industry alone “could unlock up to $60 billion in additional economic growth by 2035”, according to another study.

The recent release of the National AI Plan is intended to help capture these opportunities and to “spread the benefits” of AI across the economy – by supporting safe and secure AI adoption and skills development.

When it comes to AI’s growth trajectory, where the Australian business sector will be a year from now is extremely hard to predict. Adoption rates among businesses are not slowing down at all. The technology in its various forms – generative AI, agentic AI, multi-agent setups, retrieval augmented generation (RAG), and model context protocol (MCP) to more tightly integrate AI with existing business systems – is becoming more deeply embedded and ingrained in business operations and core functions.

It is easy to get overwhelmed with all the things you can do with AI. It makes complex concepts more accessible, and time-consuming tasks more easily executable. As just one example, you can use Copilot to analyse spreadsheets without understanding how to write SQL or Python queries, or even the basics of statistical regression analysis. And that’s just the tip of the iceberg.

My view is that AI’s potential remains limitless: with hyper-productive workforces, smarter decisions, and unmatched customer delight. But to achieve all of this it’s more important than ever for businesses to strategically partner on building a framework in order to maximise these potential benefits and opportunities.

Starting the AI growth engine

Some 60% of Australian small-to-medium enterprises will adopt AI by sometime in 2026. However, their experiences will vary significantly depending on how much effort and focus they put into the foundational elements required for AI success.

These boil down to three things: having access to aggregated and prepared data, having the right infrastructure to host AI workloads, and being able to integrate and optimise the AI for the specific context of the business.

Buoyancy for AI adoption needs to be carefully balanced against building the base capability in the organisation. Without the ability to access your data or affordable IT infrastructure to run AI in production, an organisation’s AI aspirations will remain just that: as aspirations.

So, this coming year is about getting the fundamentals of AI adoption right once and for all.

That starts with data access and data quality, which continues to be a persistent barrier to AI aspirations. This requires clear and suitable governance for both data and how you use AI. One of the advantages of current AI technologies is that it can accept and make sense of both structured and unstructured data. However, this can also quickly expose poor information management practices.  The output of any AI is only as good as the data being fed to the model. AI gets its business context from this data, and if this context has gaps, then so will the response. To achieve a focused, cogent and consistent response from AI requires both data access and data quality to be in order.

Similarly, businesses need to get their IT infrastructure in order to be able to meet their insatiable appetite for AI. For most businesses, this will mean weighing up alternatives such as hyperscaler cloud, private cloud, your own cloud and infrastructure at the network edge, closest to where raw data is being generated. A hybrid of these options is likely to be the way forward for most businesses, because it best accounts for where their data is and where is the most cost-effective place to process it using AI.

Navigating all of this is a key challenge. Finding a way – or a partner – that allows safe, secure and cost-effective use of AI can be the difference in an organisation achieving its AI ambitions.

About the author

Petar Bielovich is General Manager, Data & Analytics for Atturra. He leads a team delivering data, analytics and AI solutions, enabling digital transformation and generating more value from all forms of data. Petar has more than 25 years’ experience working with clients, including Australian Defence, Boral, Telstra and Nestle, and has worked for large professional services organisations and start-ups.