Chief Scientific Officer, Simon Wilson discusses where to start with Shared Mobility and Data Analytics

Shared Mobility and Data Analytics:
Where to Start

First in a five-part blog series on the part that analytics plays in shared mobility solutions

There are two fundamental reasons for collecting and analysing data, to reduce costs and to grow revenue. When it comes to the shared mobility market (car subscription, car rental, car sharing), it makes sense to be data-driven from day one because operators are often experimenting with nascent business models. The quicker they can accrue information about what’s working with customers, the greater the chance of their business succeeding.

The good news is that technology is working in their favour. Some of the most valuable data is collected from the growing number of sensors in cars, fuelled in part by the shift to assisted and autonomous driving. Not so long ago, collecting, storing and analysing data from moving vehicles was a big challenge, but thanks to the combination of faster mobile networks (particularly 5G) and developments in cloud computing, it’s a much easier process.

There are essentially three sources of data of interest to shared mobility companies:

1. Customer Data

As with every business, having the ability to profile and better understand customers is key to keeping their business and persuading them to spend more on additional services. It starts when they sign up, but with the right systems in place it should extend throughout the customer lifecycle. At GTS, we talk a lot about the lifetime value of customers because our Car-as-a-Service platform has multiple data points for forging a better understanding of their experience and changing expectations over time.

2. Car Data

The combination of in-car sensors, mobile networks, and cloud computing allows OEMs and dealerships to capture and scrutinise multiple aspects of vehicle performance. Telemetry is not new, but more advanced data collection lets you monitor and measure everything from distances traveled and fuel consumed to the technical performance of individual car components. Driver behaviour can also be analysed, from in-car entertainment preferences to how safely they drive, providing valuable insights that will inform the customer relationship.

3. Third-Party Data

Publicly available data, such as a population census, is often free and useful for analysing local demographics. Finding an area that has a high percentage of professional 30-somethings, for example, might minimise risks around where to launch an urban car sharing scheme. National databases can inform global strategies, pinpointing countries where there is a shift away from car ownership. Paid-for surveys might also help narrow down options or shine a light on limitations in public transport, highlighting opportunities for setting up a low-cost car rental alternative.

In my role as Chief Scientific Officer at GTS, I have helped the company develop a broad range of analytics services, leveraging a wide variety of statistical methods – some are foundational and go back hundreds of years; others take advantage of the latest AI-powered tools that leverage LLMs (Large Language Models). Advances in compute power and storage has been the game changer, enabling methods such as deep learning neural networks to do the heavy lifting.

Whatever the method, elegant statistical methodologies or industrial strength computing, common pitfalls remain. You have to learn to work with missing or bad data and understand the significance of outliers. The same principles apply, it’s just the scale that’s changed. What technology brings is accelerated access to different models at an affordable price, which is why data analytics should be part of the toolkit of every aspiring shared mobility business, and why it’s integral to GTS propositions.

Next blog in this series : Using data to reduce CaaS costs

By Simon Wilson, Chief Scientific Officer, GTS, and Professor of Statistics at Trinity College Dublin