Food delivery apps ride on power of data

Food delivery apps ride on power of data

Data analytics is increasingly being used to stay competitive in the local food delivery scene, with apps like Deliveroo and Foodpanda leveraging it to boost manpower efficiency and even dream up brick-and-mortar concepts. 

SINGAPORE: With just a few taps on a screen, piping hot meals from hawker stalls to fine dining restaurants can arrive at customers’ doorsteps in minutes.

With the food delivery scene heating up, industry players such as foodpanda and Deliveroo — which capture the lion’s share of the market — are increasingly tapping on data analytics to better compete and cater to customer tastes.

According to managing director of foodpanda Singapore, Mr Luc Andreani, data currently drive about 90 per cent of its decisions, such as its rider-order allocation system. The platform has shifted from manual to automated processes as it scaled up over time.

“At the very the beginning, we recruited riders and allocated them to a given order. So an order would come and manually we'd decide, OK this order goes to this rider,” Mr Andreani said. 

“We fully switched a few years ago to a fully automated system, which is based on an algorithm that … calculates tens of thousands of operations every second."

The system helps to "optimise the supply on one hand – meaning the riders on the street – and demand on the other hand – meaning the location of the customers", showing the best path for the rider to follow, said Mr Andreani.

This has helped shave delivery time by half to around 30 minutes, and tripled manpower efficiency, enabling its fleet of 3,000 riders to deliver three orders per hour on average.

Since foodpanda launched its central kitchen and dine-in space in Woodlands in March, the platform has shared weekly data reports with restaurant partners, including recommendations for under-performing brands.

The Berlin-based player also has plans to launch a new app feature in the second quarter of this year, offering users a customised list of restaurants based on past orders.

Another major player to incorporate data insights is Deliveroo, which set up its second kitchen concept, dubbed Deliveroo Editions, in Lavender in April.


"We've taken a look at the data of customers within this neighbourhood and adjacent neighbourhoods, to determine what are the most popular cuisines, and also which cuisines we should add to the area," said Deliveroo CEO Will Shu. 

"So in that case, restaurants know ahead of time if what they produce is something customers will really like."

The Deliveroo Editions 2 site also has a "click and collect" option to help improve delivery efficiency. In a worldwide first for the London-based platform, customers can place their orders online and pick up their meals without having to pay a delivery fee.

Data collection and algorithms are also used to provide real-time feedback to restaurant partners, including key peak times, to help improve delivery efficiency.

The platform is also partnering seven restaurant vendors at its Editions sites to create virtual brands and exclusive menus based on data analytics. 

One of them is VIOS by Blu Kouzina, a virtual spin-off from Greek restaurant Blu Kouzina. Since its launch in 2017, the virtual brand has made use of daily order reports to help cut down food wastage by up to 15 per cent and tripled its revenue.

It also cross-shares around 65 per cent of its ingredients between its two menus, which has made its operations more "effective and efficient", said its co-founder Dennis Tsakiris. 

"The two menus side-by-side complement one other – if there's a dip we could use in the Blu Kouzina menu, let's say a yoghurt dip sauce, that could also be used as a small portion also in the VIOS menu."

Source: CNA/aa(hm)