Download Hi-Res Image • Merage School Operations and Decision Technologies Professor Rick So • Photo by Issac Chan

Better Customer Service and Increased Income for On-Demand Businesses

July 18, 2018 • By Laurie McLaughlin

On-demand business platforms provide a wide range of services – food delivery, transportation, pet care and more – and routinely contend with the supply-and-demand balance between the availability of independent service providers and the needs of impatient customers counting the minutes as they wait for their dinner, a ride or a dog-walker. The authors of a new study suggest offering independent service providers a combination of time-based pricing and variable payout ratios that fluctuate according to demand could result in quicker service for customers and increased income for both the service providers and on-demand business platforms. 

On-demand services have long been part of the business landscape. In recent years, the sharing economy – a revitalized marketplace of goods and services that are utilized or borrowed without having to purchase the product or means for the service – has grown quickly. Many of these services are well known, such as Uber and Lyft (ride-hailing services), Blue Apron (meal delivery), Google Express (grocery delivery), GrubHub (delivery from restaurants) and Wag! (pet-care service).

“To meet dynamic customer demand anytime/anywhere, it is economical for on-demand service firms to use independent providers (or agents) to fulfill customer requests quickly. However, using independent agents to deliver on-demand services can be challenging, as work participation of independent providers is primarily driven by earnings,” say the authors in the study, “Coordinating Supply and Demand on an On-demand Service Platform for Impatient Customers,” to be published in a forthcoming issue of Manufacturing and Service Operations Management. “As independent agents do not get compensated for idle times, earnings depend on wage rate and utilization, whereas utilization depends on customer demand, … [and] customers depend on two key factors: price and waiting time.”

The authors look at the need for a profitable balance between the number of providers (Uber drivers, for example) available and the demand from customers (people who need a ride) from the point of view of an on-demand service firm or business platform (which arranges for the drivers to pick up the passengers). The authors suggest that in addition to adopting time-based pricing – a pre-set price structure that rises and falls with demand – the firms can also use a variable pay-ratio structure that increases or decreases the percentage of the fee that independent providers receive from the business platform according to demand load.

For example, Uber pays independent providers 65 percent to 80 percent of the payment received by the customer, but determining that ratio isn’t necessarily used to regulate – or incentivize or disincentivize – the number of drivers available. It’s best to have fewer willing drivers during slow times, so that those who are available can be better utilized to match the low demand; and of course, it’s best to have many drivers during busy times. “We argue that it’s actually better, if possible, to use a different payout ratio at different times, which we find is more valuable to both the platform and the service providers,” says one of the paper’s authors, UCI Merage School Professor of Operations and Decision Technologies Rick So. “During peak hours, the platform should lower its commission rate and entice more service providers to participate to serve the high demand. During the low-demand period, they should increase the commission rate, which of course, lowers the payout ratio, in order to reduce the number of service providers during that time.”

So contends that the reduction of commission for the firm/platform during peak demand would capture more potential customers with the increase of available independent providers serving customers who know they may rely on the business to deliver quick service when they are in a hurry. “As a result, while the platform receives less commission with each transaction, they are able to serve more customers and profit may actually be higher,” says So. “We argue that time-based pricing in combination with a flexible payout ratio should be considered and more readily adopted in practice.”

Adoption of both time-based pricing and variable payout ratios and ensuring a supply-and-demand balance requires companies to have a deep understanding of their usage data. “By understanding the demand pattern, the platform is able to forecast accurately what the demand might be at different times of the day,” says So.

For this study, researchers specifically examined their model from the point of view of a single platform – a business with a monopoly on a particular market. “But, in a market where you have strong competition, you also have to react to what other platforms are doing in competing for both the service providers and customers, and we are looking into how that would change the dynamics,” says So. “Our future research will examine how platforms can compete against other businesses for service providers.”

The study’s authors are Merage School PhD alumnus Jiaru Bai of Binghamton University School of Management, Rick So of the UCI Paul Merage School of Business, Chris Tang of UCLA’s Anderson School, Xiqun Chen of Zhejiang University’s College of Civil Engineering and Architecture and Hai Wang of Singapore Management University’s School of Information Systems.

Rick So is a professor of operations and decision technologies at UCI Paul Merage School of Business. His research expertise is in the areas of operations management, supply chain management, design of production and service systems, and business process management. So has published more than 45 research articles in major academic journals and has presented his research at conferences worldwide. He was a recent recipient of the Excellence in MBA Teaching Award, Senior Faculty Research Award and Faculty Service Award. So served as the associate dean of undergraduate programs from 2009 to 2012 and from 2014 to 2016. Previously, he worked at the AT&T Bell Laboratories, where he served as an internal consultant to various AT&T manufacturing facilities. He also served as the head of the department of management at the Hong Kong Polytechnic University and held visiting professorships at the University of Hong Kong, International University of Japan and the National University of Singapore. So earned a BS in mathematics and computing and information science from Roosevelt University and MS and PhD degrees in operations research from Stanford University.