Realising the true value of an organisation’s data.

By Kallol Dutta, Spark Business Group Data and Automation Lead, and Evan Wilson, Qrious Head of Customer and Advisory

Organisations have been collecting and collating data for years but many have missed the opportunity to realise its true value.

While technology has played its part in data monetisation – first cloud and now AI – the failure to enable it effectively often lies in other areas, such as business alignment, culture, and governance.

Which is why creating a robust strategy that considers an organisation-wide view, will pay dividends further down the track.

The first move, however, is to decide what path your organisation will take to data monetisation. It can go in the direction of external monetisation (selling data products) or it can focus on internal monetisation - using data to improve customer value and boost business efficiency in a way that offers a measurable return on investment.

At Spark, we favour the latter road – it can take longer but ultimately leads to greater returns, as demonstrated in the development of a comprehensive AI and data programme at Spark.

The first was known as BRAIN, which stands for Build Robust AI for Next-Best-Action. It is designed to help improve our product offering by sifting through a vast trove of data points.

Our next major initiative at Spark is also well under way with the use of generative AI across the enterprise. In the era of generative AI, data isn’t just advantageous—it's vital to enable both predictive and generative AI use cases to deliver better customer experiences.

We began creating BRAIN four years ago to address the challenge that at any point in time, only 1-2 per cent of the addressable market is shopping for a plan. The idea was that if we could predict who those people were, we could send them the right offers at the right time.

Over the last three years, Spark has delivered a 17 per cent efficiency gain in our marketing investment annually through our BRAIN capability. The following five steps to achieve this – vision, organisation, governance, technology, and culture – can also be followed by your organisation.

1. Vision – a data product to support business strategy

It can be tough to know where to begin, which is why the first step to meaningful data monetisation is having a vision. What we mean by this is twofold. Firstly, it’s about having a senior person, or people, as evangelists for data and AI in alignment with your business strategy. An organisation will struggle to progress a data monetisation strategy if it doesn’t have buy-in – and preferably enthusiasm – from the very top.

Secondly, it’s finding the right data product. This can be achieved either by going bottom-up – looking at the data you have to determine potential products, or top-down – identifying a business problem or opportunity that delivers on the business strategy and then finding or collecting the right data.

It’s also about asking the age-old question – what problem will deliver the biggest return on investment. It’s critically important the problem has a clear link to the business strategy to deliver significant returns.

It can be useful to have an external data expert help your team brainstorm new ideas and consider which are most achievable. They will provide an unbiased perspective and a structured approach to identifying, prioritising and refining your data products. It’s also worth cultivating a test-and-learn process when experimenting with data monetisation, so you can fail fast until you discover the best idea to proceed with.

In addition, from the outset, think about designing for a ‘system solution’ rather than a ‘point solution’. That is, create a technical capability that is scalable, rather than one that can only be used to solve a specific problem at a specific point in time.

2. Organisation – gathering the right people

Think carefully about who in your business you need on board to help design, create, deploy and adopt a data product effectively. Look across the organisation and get those people on board early. It won’t all be technology folk; if you are looking at improving efficiency in customer service, it is critically important to involve your people in customer-facing roles.

Breaking down silos in your organisation and getting people with diverse skillsets to work collaboratively will really pay off when it comes to building an enduring data product that can scale and deliver the envisioned benefits.

3. Governance – putting guardrails in place

Consider governance structures, because this will help overcome one of the biggest impediments to data monetisation - fear. Putting governance structures and guardrails in place will help overcome the fear that deploying a data and AI solution will unleash a whole lot of trouble, for example, a privacy breach or a cyberattack.

We design data products in line with GDPR guidelines, which must be followed by any organisation doing business in the European Union and is arguably the most robust privacy and security law in the world. We also developed and published a set of AI Principles that guide the development of responsible and ethical data products.

Being confident the right privacy and security measures are in place is also hugely enabling because it forces you to think about things from the customer’s perspective. For example, most people will think it’s acceptable for Spark to know if a customer subscribes to Spotify and to act on that knowledge. They would not, however, like the idea of Spark knowing what music they listened to and then acting on that.

4. Cloud-first pays dividends 

Technology modernisation is a key step because, to enable a data monetisation strategy, you need the right foundation. Cloud provides access to technology that improves the efficiency of internal teams, giving them bandwidth to deliver new data products. Cloud also offers access to new technology that unlocks value from existing data; examples include new data science libraries or Large Language Models.

Many organisations will be grappling with some legacy technology, so it may not be easy to bring all  data contained in disparate systems together in one cloud solution.

If that’s the case, think carefully about what data will be the most useful to create a data product and prioritise that when investing in systems and solutions.

5. Culture – data is a team sport

If you have followed all the steps above, and created a product within a scalable system, then servicing your data product should be an ongoing commitment. It will evolve and change and, as it does, you will need many people within your organisation to champion its development.

Creating a culture where people are keen to learn and experiment with data and AI not only pays off in terms of better products and services, but also helps retain people who are intellectually curious.

These five steps established for BRAIN have also put Spark in good stead to move quickly in our adoption of generative AI. We have been exploring generative AI to enable our people, including customer service teams, to be more productive and augment their skills.

One of our generative AI products is BRAVO (Bold & Revolutionary Adoption of ChatGPT in Spark Via Open-AI) – our new search engine using  AI to surface answers that help our customer service advisors with custom queries and decision-making. This allows our people to answer customer questions promptly, comprehensively and accurately, with references, to provide a more positive experience for our staff and customers.

In the end, whatever product or service you create, your business will be in good company.  After all, over half of those surveyed in McKinsey’s State of AI in 2023 have adopted AI. According to McKinsey, those going ‘all in’ with AI are using the technology principally to create new business ideas and new sources of revenue, not for cost reduction. It’s on the rise in every sector.

 To join in, you just need to take the first step.

Read more on Insight Engine

Find more fresh insights on the technology-related business priorities shared by Aotearoa’s business leaders on Insight Engine:

Creating seamless customer experiencesCreating seamless customer experiences - A data-driven approach: How to tackle this widespread business challenge.

Making the most of core customer metrics: Balancing efficiency metrics with those that support customer relationships.

Kallol Dutta leads the strategy, governance, and enterprise adoption of data, AI, ERP, and automation capabilities across the group. With over 20 years of experience leading consulting, IT, digital, and data teams, Kallol has a proven track record of developing and implementing IT and digital strategies that generate commercial and customer value, while evolving Spark into a data-driven organisation.

Evan Wilson is the Head of Customer and Advisory at Qrious, where he is responsible for the teams that help our customers imagine and design data-centric solutions to industry problems. Evan has extensive experience delivering strategies and pragmatic solutions to address business problems and unlock opportunities.