Automated Analytics was established around ten years ago by CEO Mark Taylor, previously a marketing executive at companies including Hilton Hotels, Halfords and Virgin Active. Based on his extensive experience, Mark saw an opportunity to improve the customer sales journey, particularly when it involves both an online and offline component.
While the volume of online sales has rocketed in recent years, customers may start their journey online but call in with queries and special requests. That might be to try and book a last-minute vehicle MoT that doesn’t appear on the online calendar or an extra room on a package holiday. However, as Mark explains, that makes it difficult to track and understand the end-to-end customer journey.
“If you're not able to attribute the invoice value to the original click, you could be optimising your online media to do exactly the wrong things," said Mark.
"If you're a service-based business, we can help you understand which keywords are driving high converting sales, which is the trick of advertising really.”
Mark and his team developed a solution, using in-house expertise and some third-party software. As a result, the company has grown impressively – with some key clients including British Gas, Europcar and Connells.
But in an increasingly data-driven industry, Automated Analytics wanted to stay ahead of the curve and reached out for additional expertise through what was the Sheffield Innovation Programme (SIP) – a regional initiative between the University of Sheffield, Sheffield Hallam University and the Sheffield City Region Growth Hub. The Programme helped stimulate business growth and promoted the development of long-term relationships.
Unrivalled AI expertise
Through the Programme, Mark and team collaborated with Dr Chenghua Lin, Senior Lecturer in Natural Language Processing in the Department of Computer Science at the University of Sheffield. Chenghua has since taken up a position at the University of Manchester, but remains at the University of Sheffield on a part-time basis.
“There are very few people in the world that truly understand AI to the degree that Chenghua and team do," said Mark.
"It was enlightening working with them and there were certainly a few wow moments, seeing what was actually possible.”
Together, the team were able to develop a completely bespoke transcription solution to analyse and interpret customers calls, drawing on techniques including deep learning (a form of AI) and natural language processing (NLP). The resulting solution can analyse ten thousand calls in 0.2 seconds, predicting the outcome of those calls at a 91% accuracy rate.
“The great thing about working with the University is that they can immediately understand what we're looking for, and really cut through the noise and actually deliver," added Mark.
"Yes, we’d developed an AI solution in-house previously, but to really take it to that next level we needed to get outside support. It moves us into a completely different arena.”
Delivering for clients
Mark estimates that the software solution developed through the collaboration could save his company £1 million pound a year, relative to if they had continued with a third-party, commercial platform. In addition, while the field is constantly evolving, Mark also says that the project has enabled them to stay ahead of the competition in terms of technology by around two and half to three years.
Most importantly, they have seen a tangible impact using the new software with their clients – with one client in particular seeing a doubling of online revenue within three months, going from £600,000 to 1.2 million.
The relationship between Automated Analytics and the Department of Computer Science at the University of Sheffield has also evolved, with the initiation of a Knowledge Transfer Partnership (KTP), working in the same general area, but on innovative new solutions.