Keep up with the ever-evolving AI landscape
Unlock exclusive AI content by subscribing to our newsletter!!
Using Artificial Intelligence to Streamline Customer Services for Train Operating Companies
November 5, 2015
Train Operating companies and the rail industry in general are never far away from headline news, partially due to a growing reliance of the public on trains for commuting and leisure.
Statistics from the Rail Delivery Group report that on average 1.65 billion passenger rail journeys were made in the last 12 months, compared to the 801 million journeys that were made in 1997, a significant increase. The report also states that passengers make on average 24.7 journeys each year, a 60% increase on 1998’s figures. With increasing numbers of passengers, the expectation to offer great customer service is paramount to the success of train operating companies.
Now more than ever, customers are becoming more vocal and creative in the ways in which they communicate with train operators, through traditional channels such as telephone calls and letters, through to emails and social media. Due to this influx of unstructured communications and the unpredictability of the rail infrastructure, it is important that organisations find a way of managing the plethora of data they receive from customers on a daily basis.
Although there are various trusted and established systems and processes currently available to train operators, such as Customer Relations Management systems (CRM’s) and Robotic Process Automation (RPA), these often fall short of the mark.
CRM’s are very useful for recording and storing information across internal business operations. The system seemingly works efficiently when it comes to dealing with structured data and if the system is fully integrated across all departments. CRM software falls down with understanding the content and meaning from “unstructured” correspondences (data that comes from the customer). Train operators are therefore, reliant on outsourcing the task of reading, understanding and processing these correspondences or are forced to place significantly higher demands on skilled customer relations teams. Outsourcing labour intensive clerical tasks is traditional and trusted, but can be costly, time consuming and not free from human mistakes.
In addition, in the case of call centres, the caller’s information isn’t always easily accessible on CRM systems and will add minutes to calls, due to the handler being required to ask personal details that may already be on the system. In these cases, the CRM becomes clogged by duplication and inconsistency, adding to the problem of the time taken getting correspondences to the correct business department and making the data useless.
Also on the market are RPA software programmes, which work by automating the handling of structured data between systems. In order for this data to first become structured, a person will need to have read, understood and processed it – again at time, cost and not without error.
On the whole there is not one technology system that seems to be effectively dealing with the all of the differing types of data received from customers and it is human understanding and intervention that seems to be most trusted.
Other industries crucial to the infrastructure of the country are now turning to advanced technology to assist with their customer service communications handling. Consumer expectations increase and so organisations are starting realise that they need to understand what customers are saying in real-time and respond appropriately without reverting to manual labour. Artificial intelligence (ai) and cognitive learning technologies may be one of the emerging solutions to this problem. For example the retail, logistics and finance industries are effectively using automation and artificial intelligence to handle their customer service and unstructured data. Automation and specifically cognitive learning technology enables organisations to consume and understand what their customers are saying regardless of the fact that the content is unstructured and unpredictable. More importantly, the technology is able to learn and so the work force doesn’t need to scale to cope with growth or unexpected surges in demand. For example, the cognitive learning technology could support when a company receives a surge of complaints due to a rail strike or engineering work. The system could upscale accordingly without the need to employ more people or outsourcing.
Organisations will be able to process what customers are saying in real-time and respond appropriately. An increased amount of the correspondence could be dealt with without human intervention, ensuring that customers receive a swift and consistent service. Cognitive learning technology can give meaning to even the most unstructured content. Organisations will be able to realise the value that would otherwise remain buried in unstructured content, providing insights and making sure that relevant information is shared with the right people and systems without delay. The productivity of staff increases and customer service and satisfaction improves, which ultimately leads to an improvement in reputation and financial performance.
Virgin Trains is an innovative organisation within the industry that is effectively implementing a cognitive learning technology solution to help them deal with the increasing customer correspondence received. They are committed to delivering excellence to its growing number of customers, a commitment that has become synonymous with the brand.
Virgin Trains engaged with Celaton to deploy its inSTREAM platform and apply Artificial Intelligence to streamline the labour intensive administrative tasks and decision making in handling customer emails. All customer emails are now received by inSTREAM. Unique to inSTREAM is its ability to learn the pattern of unstructured content through the natural consequence of processing it. As a result, inSTREAM is able to read, understand meaning/sentiment, categorise and then recognise key information within customer emails as they are received.
inSTREAM relies on its own confidence in processing documents and data and therefore if confidence levels are not high enough then it will refer these exceptions to the customer relations team for assistance. This human intervention helps to teach inSTREAM and enables it to become more confident with every transaction. Learning is permanently enabled to achieve continuous optimisation and will, subject to confidence, minimise or eliminate the need for human intervention. Celaton’s customer service team are continuously monitoring transactions levels from its offices in Milton Keynes and can intervene where this is a potential breach in service levels.
Despite the complexity of the content that needed to be streamlined, inSTREAM was able to deliver tangible benefits within days of go-live. The daily processing time and manual labour involved in dealing with customer emails was reduced by 85% and this has had a significant impact on the time taken to respond to customers so significantly improving the customer experience. In addition, it has freed up staff within the Customer Relations team to work on “first time resolution” and deliver exceptional customer service, rather than process correspondence.
With the plethora of solutions available it is not always easy to find the right one. It is however, clear that with the advancements in technology, artificial intelligence and cognitive learning technology, the decision is becoming more clear cut and organisations are starting to realise the benefits.
You May Also Like
Generative AI Journeys with CDW UK's Chief TechnologistFeb 28, 2024
Qantm AI CEO on AI Strategy, Governance and Avoiding PitfallsFeb 14, 2024
Deloitte AI Institute Head: 5 Steps to Prepare Enterprises for an AI FutureJan 31, 2024
Athenahealth's Data Science Architect on Benefits of AI in Health CareJan 19, 2024