The company wanted voice and text analytics to support its internal audit service line. We built a solution that utilised advanced machine learning and artificial intelligence to analyse thousands of minutes of customer calls. 

Fast facts

 purple pin icon

UK financial services company specialising in insurance and asset management

Purple icon depicting a handshake

Founded in 1857 with several mergers and acquisitions over the years

purple icon depicting a leadership chart

One of the UK's largest providers of life insurance and pension services – managing over £300 billion of customers' assets

tree icon

Strong commitment to sustainability with goal to become a net-zero business by 2025

icon showing inclusive culture

Several awards for its customer service and recognised as one of the UK's best companies to work for

The challenge

Using technology to transcribe and audit calls can enable firms to analyse 100% of calls, but the challenge is uncovering deeper business insights from recordings, identifying priorities for further analysis, and developing a strategic approach to apply it. The company needed support developing voice and text analytics capabilities to tackle this obstacle and build a cohesive system to monitor calls and explore options for expanding these capabilities into other business areas; including first and second line, data analytics, data privacy and protection.

How we helped

Starting with a small dataset, a proof-of-concept was developed to analyse the initial set-of-use cases. This was then incrementally built upon with a larger and more targeted dataset. Regular meetings were held with the business to understand the most valuable key metrics. Specifically, on customer outcomes, potential identification of vulnerable customers, and call-centre performance.    

The next step was assessing the facility for extracting the necessary data from existing sources. We then considered which techniques should be applied to the data to generate the metrics. Machine learning, AI, and visualisation solutions were applied where needed.

High-level architecture

The results

We developed voice and text analytics use cases across the first, second and third lines. The project’s technical environments: Azure, Python, and Power BI supported the analysis of various quantified data dimensions.

We played a key role in successfully exploring and designing a solution for providing supporting data for regulatory mandates, such as the Consumer Duty, where information on customer journeys and outcomes needs to be easily and quickly qualified.

This solution provided a broader reach on product development, identifying client priorities and propensity, and better customer journeys for their respective pension and insurance products.

About our team

Our team offers advisory, assurance, and implementation support to help you make the most of your text analytics capabilities. Working closely with your business, we can play a key role in successfully exploring and designing a voice and text analytics programme to address regulatory risk management shortfalls.

Nikhil Asthana

Nikhil Asthana