Predict risk of churn for individual customers or clients and recommend intervention strategy - this may be involuntary (ie due to bankruptcy) or voluntary (ie switching accounts) churn

Accelerate identity verification (e.g. photos, biometric reading) for new and existing customers. This enables offering of enhanced services and greater security.

Match expectations from both sides of a 2-sided online market. Typically this goes beyond simple price matching and includes a variety of other variables that may have different levels of weighting to market participants.

Advanced analytics on all customer contact data across multiple channels (Including real world monitoring) to uncover insights to improve customer satisfaction and build a holistic picture of their status.

Tailor debt collection processes by identifying which practices are most effective for different segments of customers. This is a sensitive process with occasional risks.

Call routing (i.e. determining wait times) based on caller id history, time of day, call volumes, products owned, churn risk, LTV, etc. Route calls to most capable agent available and ideally leading to fewer agent-handled calls - hopefully leading to increased customer satisfaction and reduced handling costs.

Improve IVR effectiveness through deploying voice to text and natural language processing (NLP) to better capture and enable response to customer queries. Understanding customer pre and in-call intent helps reduce time to serve and potential customer problems. This leverages Natural Language Processing (NLP) and machine learning to estimate and manage customer's intent on calls. In-call assessment enables multiple functionalities: e.g. call routing, issue triage, automated responses.

Automate customer service voice conversations through a conversational agent chatbot enabling high volume, fast reaction customer support. Unexpected questions will likely 'break' the chatbot system so consumers need to be clear that they are interacting with a machine.

Monitor overall customer contacts to identify major trends in their questions, concerns, etc.

Leverage Natural Language Processing and machine vision to build approach to automatic responses to customer requests that come through social media channels.

Advanced analytics on call data to uncover insights to improve customer satisfaction and increase effectiveness

Automate customer service text conversations through a chatbot enabling high volume, fast reaction customer support. Unexpected questions will likely 'break' the chatbot system so consumers need to be clear that they are interacting with a machine.

Understand customer sentiment and customer value to prioritise inbound emails for response and trigger potential replies.

Bots will listen in on agents' calls suggesting best practice answers to improve customer satisfaction. Putting the right data on the operator's screen to ensure they are prepared with the context of the call to speed resolution and maximise likelihood of customer satisfaction.

Analyse how customers are interacting with a deployed chatbot - typically this will generate performance metrics but also customer, service and business insights.

Predict call volume for the purposes of staff scheduling and optimisation of resources.

Authenticate customers without passwords leveraging biometry to improve customer satisfaction, reduce wait times and speed up service access, leading to lower drop rates. Key impact should be to lower fraud risk.

Using voice and text analysis to uncover overall customer sentiment - negative or positive - sometimes in real-time as displayed when they contact the company; for example through the contact centre.

Use AI to provide real time translation services. This has both B2C and B2B applications. Current depth of access on traditional language pairs - e.g. English: Spanish - are being extended to other languages, although many cross-translations pass through English.

Ensure that high-volume, low value-add end-user issues are handled with the support of AI for voice recognition / security purposes.