How BT Uses AI & Big Data to Improve Customer Experience, Forecast Call Centre Demand, & Fend Off Cyberattacks

For many, artificial intelligence (AI) is still synonymous with dystopian futures where the world is ruled by murderous sentient robots. But this characterisation is now starting to wane as AI becomes much more commonplace. Indeed, with the likes of Siri in our pockets and Alexa in our homes, artificial intelligence has become not only a household name but a household presence in our daily lives.

In business, AI is making even more of an impact, disrupting virtually every process in every industry. As Dr Robert Hercock, Chief Researcher in Disruptive Technologies at BT puts it, “Despite the hype, AI isn’t going to go away. It’s driving business transformation.” This statement certainly holds true with regards to the UK telecoms giant’s own operations. At BT, AI is used to improve its various operations in a multitude of ways – from cybersecurity to customer experience to forecasting call centre demand – and it’s big data that’s fuelling the AI engines.

Hear how other business leaders in data science are using AI and technology at Data Leaders Summit 2019, one of Europe’s fantastic data science events. Download the agenda today!

Improving Customer Experience

With 18 million customers in the UK alone, BT is not only one of the biggest and most complex telecommunications providers in the country, but it also has access to vast amounts of customer data that can be used to drive competitive advantage and improve the customer experience. However, with so much data, just making sense of it all and extracting the insights needed to realise improvements is one of the biggest challenges BT faces in and of itself.

As one BT analyst said: “I’ve got too much data at my fingertips, I need the tools to tell me where to focus my attention first.”

And this is where AI comes in. In its own network operations, BT has been utilising AI to transform how the company predicts service outages. Using anomaly detection and predictive analytics, analysts gain a prioritised view of what equipment and machinery are most likely to cause an outage, and AI tools then classify the state of repair and prioritise maintenance schedules.

Paul O’Brien, Director, Service, Security and Operations Lab at BT, describes one of the company’s recent stand-out big data customer experience projects, in which data from 20,000 BT and Openreach engineers – along with information from weather feeds, network performance indicators, workstack patterns, and more besides – is gathered for analysis.

“We then apply the latest in emerging technologies, for example, AI or machine learning, to that data so that we can predict where engineers with the right skills will be needed, and when,” explains O’Brien.

“So far, by using this technique we have improved our forecasting ability by between two and three per cent. With a workforce of tens of thousands, this equates to savings of many millions of pounds. This is a great example of BT increasing automation across its estate so that we can drive down costs, cut wastage, boost efficiency and, most importantly, deliver a better experience for our customers.”

Forecasting Call Centre Demand

BT has also started using data science and machine learning to forecast call centre demand, which enables the business to better plan its marketing activities, overall operations, and manage its long-term budgets with greater confidence. To deliver accurate forecasting, BT formed a partnership with professional services and information technology company CACI to make use of its CACI Forecaster solution.

Using Forecaster, BT is able to better understand the drivers of call centre demand, including the effects of marketing campaigns and caller information – such as where the customer is in the buying cycle, and what products or services they’re most likely to be interested in. With this information automatically processed and visualised, BT is able to offer an improved level of service, optimise planning, and enable managers to spend less money by accurately anticipating demand and allocating resources more effectively.

The result of the partnership is that BT is now able to produce 66 forecasts in the same amount of time it took the company to produce 11 using previous manual processes, and the data-driven predictions enabled by the Forecaster solution allows BT to gain greater insights with fewer call centre resources.

Cracking Down on Cyberattacks

BT researchers are also utilising AI in the company’s fight against cybercrime. The threat landscape is indeed huge for BT. For hackers, infiltrating BT is one of the biggest prizes in the world. BT provides the largest number of fixed-line, broadband and mobile services in the UK, and operates in around 180 countries globally.

Key to winning the war against cyber attackers is the BT Research Lab and Control Centre at Adastral Park near Ipswich, UK, which forms the heart of BT’s cyber team – a 3,000 strong army of cyber professionals working around the clock to prevent a major network attack. Here, BT’s experts have developed artificial intelligence to fight off mounting strikes that are increasingly hard to detect.

“AI is really important here and that’s because of the scale of the problem,” explained Alex Healing, BT’s Senior Research Manager, Future Cyber Defence. "The data is huge – we’ve got huge amounts of events over the network and we need AI to help us work out what’s most important, what’s most suspicious, and that really focusses the analysts' attention.”

On average, the team at Adastral Park deflect roughly 4,000 attacks each day, equating to about 125,000 each month. It wouldn’t be possible for humans to monitor more than a tiny fraction of BT’s network infrastructure and the attacks that threaten it – but with AI on the team analysing the data, the malicious activity can be spotted early and the source of an attack quickly identified.

Final Thoughts

BT’s big data and AI strategy are vast and varied. The company’s aim is to become more efficient in the way it uses data and information to manage its business and deliver better products and more secure services to customers. With analytics and data science permeating practically every arm of its operations, BT is able to extract patterns that give the company the valuable insights it needs to help its people do their jobs more effectively.

As Dr Nicola Millard, Head of Customer Insight and Futures at BT, puts it, “With people increasingly the differentiator, it isn’t about man vs. machine – it is man plus machine.”

You can hear Dr Detlef Nauck, Chief Research Scientist, Data Science, BT, speak at Data Leaders Summit Europe 2019, a data science summit taking place this October at the Hotel Palace, Berlin.

Make sure you don’t forget to download the agenda for the data science conference and event today to gain more information and insights.