Interview: Ian Ruffle, head of data and insight, RAC | Computer Weekly

Transforming front-end systems

While Ruffle has always ensured his technical capabilities are up to date, he recognises that the pace of digital and data innovation has quickened over the past few years.

It’s a tough challenge for any individual to keep abreast of all the changes in the IT industry, let alone someone who’s spent a big proportion of their working life in marketing. Ruffle’s answer is to lean on trusted lieutenants in his team.

“It’s important to get the right people around you,” he says. “As a modern data leader, you need to be encouraging people to be learning and growing and empowering them to do a great job. I can’t do everything across a broad breadth of disciplines.”

Ruffle says being a great boss and attracting the right talent is the way forward. “You want staff to be motivated and encouraged to feel like they can drive the agenda,” he says.

“It’s important to get the right people around you. As a modern data leader, you need to be encouraging people to be learning and growing, and empowering them to do a great job. I can’t do everything across a broad breadth of disciplines”

Ian Ruffle, RAC

“I love working with my people. If you’ve got an idea, you can sow the seed in someone, and then they come back and play their theory back, and you can say, ‘Yeah, we’ll do that’.”

One of the team’s priority projects is a transformation of the RAC’s front-end marketing technology. The company uses a legacy system with a relational data model.

“It’s very batch-based,” he says. “There are some real-time use cases associated with the system, but it’s become prohibitive from a cost perspective to do everything that we need to do.”

Ruffle says the transformation process involves moving to a new supplier called Bloomreach. This agentic personalisation technology will sit on top of the RAC’s existing Snowflake AI Data Cloud platform. His organisation will be working with technology specialist Caci to deploy the Bloomreach technology over the next nine months. The result of this work should be a modern system that offers real-time and AI capabilities.

“That is quite a major project for us,” says Ruffle. “We’re consolidating four different suppliers in total, and probably six pieces of technology across those suppliers, into one central solution. The back end will be Snowflake, and Bloomreach will be dealing with a lot of the front-end stuff, which should mean the seamless integration of data is so much easier.”

Creating great experiences

Another important area of work is boosting operational efficiency. Many of these efforts are focused on two of the biggest cost areas of the business – the call centre and traffic patrols.

A large proportion of the organisation’s enterprise information is now held in Snowflake, using a robust data model. Ruffle says in-house developed application programming interfaces (APIs) stream data on real-time use cases back to the platform and the people in the call centre.

“It’s about the ability to know someone’s broken down at the roadside and to send them a real-time communication, with more accurate updates than we were ever capable of before. We’ve revolutionised the whole dispatch process on the other side of the call centre,” he says.

“We’ve built a single screen for users, powered by the data in Snowflake and via various web services to make the call centre experience seamless, rather than having three or four different operational systems.”

Ruffle says the joined-up approach makes it much easier for operational staff dealing with customers to answer important questions, such as whether an individual at the roadside has RAC cover. Before bringing data together, staff had to log in to different systems to confirm a customer was covered. Now, staff benefit from the single-screen approach.

We’ve got a platform powered by modern technology. The focus on technology and data driving change in the call centre and at roadside operations has been immense, and it’s been the backbone to the last few years of evolution at the RAC

“We’ve got a platform powered by modern technology,” he says. “So the focus on technology and data driving change in the call centre and at roadside operations has been immense, and it’s been the backbone to the last few years of evolution at the RAC.”

Managing complex scenarios

After evaluating potential solutions, the organisation deployed its first Snowflake platform for the marketing department in early 2020.

Ruffle says the AI Data Cloud’s scalability and its integration with the company’s existing Microsoft Azure architecture were important success factors. Since 2020, the RAC has expanded its use of the Snowflake platform beyond the marketing department.

Now, in addition to providing a single source of truth for the business, the company uses the Snowflake Marketplace, which is an online platform where users can access third-party data to augment their own insights. Ruffle explains how his organisation uses this data to enrich its own insights and decision-making processes.

“There’s a whole bunch of events data that we get, which includes things as simple as bank holidays, but also includes things like Glastonbury and other events where you might see gatherings of people in specific areas,” he says.

“Having easy access to information that’s comprehensive and maintained is great, because trying to gather that data manually is almost impossible, and to get the level of detail around the scale of the event is transformational.”

The RAC has also developed its own unified insights platform. This pane-of-glass application, known as Mavis, uses Snowflake’s Cortex AI tool to summarise data insights. Agents working in the RAC’s operations hub can use Mavis to find important customer information, such as cover entitlements.

“We’ve got all the information in one place now, which is easy, so the application will help you find the customer and fully understand their entitlement, which is more complicated than you’d think. There are a lot of nuances to the cover that people are entitled to,” says Ruffle.

“It’s about being able to get that level of detail and having accurate information at your fingertips. When you get to the much more complex scenarios, we’re helping an agent to make decisions. Our agents can see the recommendations and make decisions. That type of recommendation wasn’t possible just a few years ago.”

Being an empathetic organisation

While AI is already impacting customer service, Ruffle says breakdown specialists like the RAC must keep humans in the loop. People contact his organisation at moments of need, and technology should be an adjunct rather than a replacement for human assistance.

“People can be quite panicked, they can be unsure, and I don’t think we’d ever want to take away the humanity of dealing with people at that moment of need,” he says. “Technology plays a significant part in being operationally efficient, but we need to be really sensitive about our service being authentic and human.”

For example, customers stranded at the roadside will be looking for an estimated time of arrival (ETA) for the RAC patrol team. Ruffle says his team will continue using AI and data science to hone the accuracy of ETAs. However, great assistance is also about emotion and engagement, such as prioritising service for people who are pregnant or disabled.

“The digital evolution for us will be about leveraging technology. Data science and AI are improving, and, in the future, I think technology will be front and centre for us when making choices. However, it’s crucial that our agents can manage situations effectively and be a human face to the organisation,” he says.

“Our approach should be data-led in the background when making decisions. Automation is great, but there always needs to be an ability to manage the things you can’t predict and to manage scenarios where you really need to be an empathetic organisation.”