A London-based startup has developed an automated delivery allocation service using machine learning.

On the dot, which is backed by logistics and courier business CitySprint Group has only been providing its service since 2015, but hopes to take on the likes of Amazon with a range of features built to meet the needs of customers across the UK, particularly with same-day and time-slot specific delivery options.

Image: Taher Khaliq, CTO, On the dot
Image: Taher Khaliq, CTO, On the dot

"We're in a unique position because of how you can optimise it and how we can meet customer needs," said On the dot CTO Taher Khaliq. "On the dot has 20 years of knowledge and with CitySprint being dominant in that same-day market and delivery, it means we have all this data and information that we can mine and process.

"We have the information so we actually can solve problems from the beginning."

CitySprint Group is a nationwide courier with over 20 years experience in providing flexible and fast deliveries for businesses.

On the dot uses machine learning to allocate the right delivery driver for different package deliveries, with options of same-day and one-hour time slots at the customer's request.

It also uses real-time data tracking to respond to the demand of retailers and customers. On the dot currently works with big retail brands such as ASOS, Dixons and Wickes to provide deliveries across the UK.

In addition, the startup provides a mobile app service for couriers to manage delivery slots effectively. This also has automation built in.

"We have this wealth of data and one of the key things about 20 years of data is going through it, finding the really useful data and turning data into information," said Khaliq. "So what we have done is we have built both a short-term and long-term machine learning strategy."

Talking about the technology, Khaliq explained that the process of data collection is not something that all delivery companies use effectively. "We're solving a problem across the full chain end-to-end, so it's trying to make a driver happy about what they're doing and the job they're doing and being able to be more predictable in the work they're getting," he added.

According to On the dot, retail customer ASOS reported a revenue increase of 30 percent year-on-year in part to the launch of its same-day delivery service.

"We run our own tests so when we go through our platform with the customers there's this high level of visibility and training we provide them," said Khaliq. "So we have a variety of applications and tools we've built so the customers can directly see what's coming through their applications.

"So we also have an understanding, because we have some behaviours of the customers and we already have the subset of customers from CitySprint. Using those rules, we've automated how we can allocate and how we can consolidate packages."

Read next: AI startups to watch: The hottest machine learning startups in the UK

What the future holds

As well as improving its automated processes, On the dot hopes that it will be used to build brand loyalty for businesses of all sizes.

"You could be a florist down the road, or you could be as big as ASOS, it's all about building that brand loyalty," Khaliq added. "Technology enables scaling. It means that as a controller you don't need to make 50 phone calls to couriers, all of that can be automated and allocated."

On the dot is built on a microservices architecture, helping the company to quickly be able to identify any problems and work on them and then deploy quickly.

"Another thing that we are looking into is making everything available on SaaS [Software-as-a-Service]," Khaliq added. "So you can integrate with a customer within three days to be fully up and running through our systems, this way it's not a two-month on-boarding process or anything. That's the great thing about the SaaS platform."

"There's always more to learn and we're very much aware of that, but because we have a lot of that information already and people who work with us as well, it puts us in a good stand for that compared to a native startup trying to build that relationship."