Elevating day-to-day Australian business operations with ML

Game-changing tech like artificial intelligence (AI) and machine learning (ML) can drive high value customer experiences for business, and raise the stakes when it comes to competitive advantage. Yet, even with broadscale access to these capabilities, many Australian organisations are only scratching the surface when it comes to realising their potential.

To leverage AI and ML capabilities across an organisation they must be met with a comprehensive understanding and acceptance of best practices. A recent National Pulse Report echoes this sentiment, revealing that 8 out of 10 Australian businesses are already on their ML journey and 86% consider that ML adoption is critical to accelerating business goals.

AI and ML have the potential to ensure that client’s basic expectations are met in the most efficient way, like real-time decisioning guidance on Netflix or Amazon, or quick dispute resolution on Uber. Clients expect this level of automated responsiveness, yet many Aussie businesses still struggle to realise the full potential of AL/ML-enabled customer experiences.

How can Aussie businesses leverage ML?

At our recent Thoughtworks’ Digital Leaders Forum, which focused on the convergence of AI and customer experience (CX), business leaders examined the potential of ML to differentiate at a deeply emotional level – creating satisfied, excited or delighted customers who feel understood, confident and secure. Following are some key insights regarding how Australian organisations can take full advantage of the power of ML in their day to day operations.

Realising the use of ML as mainstream

Thoughtworks’ recent Technology Radar report Vol.27 identifies that while machine learning once required large data sets and access to immense compute power to tackle complex business issues, the evolution of machine learning and the breadth of tools, applications and techniques now available is enabling organisations to do more with ML, in more places.

While it was once developed as a highly specialised technique almost a decade ago, ML is a powerful tool that can move the needle when it comes to organisational performance. With this potential, businesses should look more broadly for opportunities to capitalise on these capabilities, rather than think of them as having niche or specialist applications.

The report observes that when businesses are more open and pragmatic to ML experiences, it can prove to be transformational in overcoming everyday challenges that impact an organisation’s time, resources and bottom line – whether it be significantly improving internal operations, driving product innovation or enhancing the customer experience.

Prioritising in-house ML skills

Many firms still struggle to obtain the in-house AI and ML expertise they need to identify, prioritise and execute AI-enabled customer experiences. Too often, tech-driven Proof of Concepts fail to align with real business use cases, while key decision makers often fail to recognise the possibilities AI and ML can deliver. Their systems aren’t set up to gather the right data (or data of the right quality) to be able to deliver on AI/ML enabled experiences.

Many of these issues may be due to historical organisational barriers. Thoughtworks’ research with Harvard Business Review Analytic Services found 34% of business leaders believe organisational or team siloes are a major challenge for implementing a CX strategy.

Another challenge is customer hesitance. While they may expect frictionless experiences, they are not always willing to exchange their data – for example, the Capterra 2022 report finds 43% of eCommerce shoppers prefer to checkout as a guest. That means you’ll need to rely on anonymised channel, location and behavioural data to guide their experience.

Avoiding applying ML in silos

Organisations may have great ideas on how AI or ML could enhance customer experience, but the lack of in-house capabilities to evaluate, prioritise and test those hypotheses. One solution to this problem is to create focus areas of specialist expertise, which can be deployed to support teams to run small experiments that enable testing and learning.

These specialists need to be embedded within teams who define and deliver the work – not separated in a centre of excellence. Otherwise, every data handover has the potential to add friction and reduce data quality. When this approach is executed successfully, you can quickly iterate, learn and apply the same approach to the next product team platform.

Utilising ML to augment – not just automate

The Thoughtworks’ Technology Radar report foresees machine learning adoption being a significant catalyst, to augment human interactions, rather than replace them. It is worth bearing this in mind with high stakes or high stress customer experiences – such as travel disruption or a home loan approval. if the only customer experience channel available is delivered by an algorithm, there is a risk of frustration or even reputational damage.

There have been some cases of algorithms in financial services, healthcare, policing and recruitment discriminating against people on the basis of race or gender. Dr Catriona Wallace said there is much to be done to ensure AI is deployed within an ethical framework. Her co-written book, Checkmate Humanity, outlines the first taxonomy of AI harms and offers recommendations to guide organisational decision-making and get the best from this tech.

Ultimately, ML is just that – it is constantly learning, and there is a potential cost when it gets things wrong. Even the best system will occasionally fail, which is where human insight and allowance for intervention will continue to be essential. For this very reason, it is important that Australian organisations start investing in developing the artificial intelligence and machine learning skills they will need for tomorrow while adopting these technologies today.

Armed with the right capabilities and data platforms that build in privacy by design, we can find new ways to safely deliver experiences that customers expect – experiences that make them feel delighted, satisfied and understood. Since ML is highly dependent on the quality of training data, the standard cautions remain on vulnerabilities and inherent bias in data sets.

While there are multitudes of open source tools that help interpret machine learning models and provide some transparency you still need skilled people to interpret the results and make decisions that are aligned with the organisations and societal values.

As computational power grows on devices of all types, and with the world becoming increasingly interconnected, open source tools will continue to become more prevalent and easier to use, evolving ML into accessible technology for even the smallest organisations.

With more stringent requirements and awareness around privacy and personalised information, techniques such as federated machine learning are ensuring greater privacy for sensitive information in internet of things (IoT) and mobile use cases.

Unleashing the full potential of ML by embracing the tech and best practice, particularly leveraging data to deliver best-in-class customer experience, is a vital way Australian firms can better compete on the world stage and set their path to future growth for tomorrow.

Meg Blake is the Product and Experience Design Principal at Thoughtworks Australia whereas David Colls is the Director, Data and AI practice at Thoughtworks Australia.