The sector needs to up the pace on harnessing artificial intelligence

Our first major step saw our in-house team build a cost-effective predictive AI model, bringing together existing service data, including repair history, customer contact patterns and known vulnerability indicators. By analysing information we already have, the model predicts the likelihood of future complaints where it sees a high risk of dissatisfaction amongst customers.

This has helped our customer-facing teams better focus their attention where it’s needed most, enabling them to intervene earlier and meaning customers receive the right support to resolve their issues. The outcome of those high-risk cases: a customer satisfaction score of 9.9.

“AI must be seen and used in ways that strengthen and support human decision-making, keeping human judgement at the centre of the services we provide”

In reflecting on our work to date and in considering the lessons learnt, we’ve recognised that several key principles must underpin how AI is harnessed when seeking to drive improved customer experience.

AI must be seen and used in ways that strengthen and support human decision-making, keeping human judgement at the centre of the services we provide. For teams to properly trust AI and to willingly use its outputs in helping manage complex cases, such as in our work, plain-English explanations are vital. These are far more helpful than teams being presented with complex scores or highly technical outputs.

Whether it’s the complaint-risk model we’ve developed or others where AI can intelligently aid decision-making, the key is embedding AI within teams and embedding it well. We’ve learnt this matters more than the technology itself. In our view, AI works best when it supports real teams and real workflows and when there’s clear accountability.

We can confidently say that, while our own progress has been rapid and ambitious, it’s also been well measured and carefully governed. Every step we’ve taken has been done so with a strong emphasis placed on control, oversight and responsibility.

All developments remain fully aligned with key regulatory frameworks, including the EU Artificial Intelligence Act and GDPR, embedding compliance and ethical considerations into our processes from the very start, rather than as an uncomfortable afterthought.

At the same time, we know that innovation carries broader responsibilities that outstrip regulation alone. It’s for this reason, that we’re actively monitoring, assessing and working to reduce the environmental impact associated with our AI initiatives.

By combining speed with discipline, and innovation with accountability, we’re building AI capabilities that are not only effective and forward-looking, but also sustainable, transparent and trustworthy in the long term.

“If we don’t properly realise the gains and make the most of the very latest our evolving technological landscape offers, we’ll be doing our customers a clear disservice”

But if we’re to see AI and data properly employed and maximised for customers’ benefit, we need them placed at the centre of our organisations. To do this, we need a cultural step-change, so let’s see our sector up the pace, given how well businesses working in other areas are already embracing AI to deliver better customer experiences.

On this basis, we should also recognise the opportunity to learn from others as we start to catch up with those businesses that have already seen positive customer outcomes. As a sector, collaboration and knowledge-sharing has always been one of our strengths – let’s capitalise on this to help us make the rapid strides we need to make.

If we don’t make the most of the very latest our evolving technological landscape offers, we’ll be doing our customers a clear disservice. The long-term benefit to our organisations and crucially, to all our customers in making such a step-change is significant. Predictive customer support will soon be an expectation, not an innovation.

Fiona Harris, chief information and data officer, VIVID