By Dan Somers, CEO of Warwick Analytics
More banks are turning to practical AI to rapidly analyse customer conversations for sentiment and emotional intent to get the insight and automation they need to transform their customer service and operations.
Here we look at 5 ways in which banks are using AI to process their customer feedback more effectively:
Processing incoming queries more efficiently
AI can remove the need for manual review of each incoming query and enables banks to handle them effectively from the outset.
The analytics can facilitate a much smoother omni-channel experience for the customer by: identifying which channels your customers are best suited to – and which work best for specific types of interaction; understanding the causes of channel failure and what drives customers to switch; and reducing customer effort by delivering service in the customer’s preferred channel first-time.
As a recent example, at one bank we were able to reduce the maximum time to respond to a customer from 3 weeks to 5 days. The solution used AI and machine learning to automatically analyse and prioritise all customer emails in near real time and routed high-priority cases to a dedicated work queue for fast action.
Automatically identifying customer intent and emotion
When different people are voicing different issues, they will use different words and sentiments. Vital data is often missed with traditional models and manual processes. For example a customer at a bank might say ‘by the time they called back, the bank was closed’. The keyword would be flagged as ‘closed’, when in fact the main issue was the call back. There are also other limitations with using just keywords such as sarcasm, context, comparatives and local dialect/slang. The alternative is to analyse text data using ‘concepts’ instead of ‘keywords’. This can be done effectively with AI.
Fast tracking customer complaints and issues
With AI you can send complaints straight to the relevant team for a faster resolution. We’ve helped banks reduce resolution time by up to 3 days which really boosts customer retention.
Dealing with specific complaints manually involves using more and more case handlers. Routing complaints automatically and prioritising by issue and category is also difficult due to the nature of complaints i.e. unsolicited, long and sometimes multi-topical. As a result, manual classification is often impossible within an acceptable time frame for the unhappy customer.
Using the latest AI however, banks are now automatically classifying unstructured data to provide an early warning of issues that need resolving fastest. This can lead to better and quicker outcomes at a much lower cost.
Spotting vulnerable customers early
Under the Financial Conduct Authority (FCA) front-line staff need to be able to spot different types of vulnerability in customers and support them accordingly. However, the volume of communication is just too much to carry this out manually.
The latest in AI speech transcription and text analytics is able to automatically detect hints at vulnerability from conversations with customers. The conversations are automatically analysed by to detect emotionally-driven comments that indicate vulnerability such as a basic lack of understanding, likelihood of a disability and circumstances. These vulnerabilities are flagged to the relevant members of staff for action. Regulated firms can also accurately understand the drivers behind the vulnerabilities so products, services and communications can be reviewed accordingly.
Banks using AI during Co-vid 19
During Co-vid 19 many banks have customer service agents working from home and/or in strict shifts. There has been a move from voice to webchat for many to cope with these changes which brings its own challenges and opportunities. Post-C19, many of these situations are expected to stay in place or at least not revert 100% back.
AI is helping to serve customers better focusing on taking cost out whilst keeping CSat up and channel switching down by improving chat optimisation, email, complaint handling and chatbot supervision.
Case study: Improving customer loyalty
A major UK bank was looking to improve its customer loyalty. It was already using the latest
analytical tools including social listening, sentiment analysis and a large data science team
but they were experiencing limitations and not making enough progress. They were also interested to see what online feedback their main competitors were receiving.
A number of key recommendations for the bank were identified using AI analysis:
- A 10% increase in CSat (c. £200m pa revenue) from operational improvement
- Comparable best-in-class churn e.g. Nationwide is 25% lower
- Online and mobile banking is a key issue, and is causing direct churn
- Drivers of churn are mostly customer service, branch closures, marketing offers, interest rates and vulnerability issues
- Early warning can help predict churn tactically and intercept likely churners
- 28% of Tweets and potentially all non-voice queries can be automated. This could be a £20m pa saving
- Business banking, current accounts and ancillary services have the highest churn, and insurance the highest negative advocacy
- Mortgages, current accounts, savings and overdrafts cause the most attritional set-up
- There are distinct patterns and opportunities to adjust customer services resources to reduce churn and costs
With AI, this level of insight can be set up in a matter of days, delivered in near real time and without the need for a data scientist to maintain the model.
