Subscribe
About
  • Home
  • /
  • CX
  • /
  • Uber powers SA contact centre investigations with AI, ML

Uber powers SA contact centre investigations with AI, ML

Sibahle Malinga
By Sibahle Malinga, ITWeb senior news journalist.
Johannesburg, 18 Apr 2023
Inside Uber’s Cairo-based contact centre.
Inside Uber’s Cairo-based contact centre.

E-hailing company Uber says its $20 million (R366 million) Egypt-based contact centre is constantly augmenting its investigative technologies to resolve complex queries from South Africa, through an omni-channel strategy.

ITWeb was invited for a behind-the-scenes tour of the US-headquartered multinational e-hailing firm’s Cairo-based contact centre. The Centre of Excellence deals with queries ranging from crime incidents, to abusive passengers and car accidents across 14 countries, including SA, in the Europe, Middle East and Africa region.

During its launch in 2017, Uber announced it would invest R366 million over five years in its second-biggest contact centre globally, to enhance customer service throughout the region. It has 345 employees who support the centre’s more than 20 lines of business − including safety and security, rides, Uber Eats, etc − in four languages: Arabic, English, French and Urdu.

Artificial intelligence (AI)-enabled analytics, robotics, remote desktop technologies and call centre software are among the tech improvements the company has introduced to provide its riders, drivers and couriers with fast and efficient customer service support across various channels, including e-mail, in-app support, web chat and phone calls 24 hours a day, seven days a week.

Uber MEA customer experience lead Mohamed Nasif told ITWeb the safety agents who handle a large number of SA’s safety incident reports rely heavily on advanced technologies to investigate and provide fast, accurate and empathetic responses for every reported safety incident.

“In cases of violent incidents, it's important that the agent moves very quickly and efficiently. The process starts with identifying the type of incident, which is conducted using a machine learning system that uses keywords to classify the type of incident being reported.

“Each process is technology-dependent, and each agent has an advanced set of tools which they use to try and understand the incident that has happened and how to provide the best support, through gathering robust data and insights to investigate, resolve and enable future incident prevention.”

Research conducted by security company 0861 Clearview, lists SA as the most dangerous country in Africa, in terms of the high level of violent crimes.

The past few years have seen e-hailing drivers hold nationwide protests, urging government to intervene in the increasing incidences of crimes they are subjected to at the hands of criminals and rival metered taxi operators.

Since inception, the industry has been tainted by incidents of violent attacks, and in some instances, drivers being hijacked or killed.

Uber’s contact centre receives a significantly higher number of sexual and physical assault incident reports from SA compared to other countries – which prompted the introduction of its audio-recording feature in SA, the only country where the service has been introduced so far.

Mohamed Ibrahim, Uber Centre of Excellence senior community operations manager, pointed out that resolving safety incident reports can take anything from a few minutes, to several days, depending on the nature of the incident.

“Our in-app safety features are tailored according to societal needs of riders in that particular region,” he explains.

“The audio recording has not only assisted in post-incident investigations, but has also assisted in making users alert with their behaviour, as it is always recorded.

“The core competency of the agents is critical, and the selection process of the team itself is very rigorous, to make sure they have the required competencies. Background checks and assessments of different types are done to ensure they have the right profile of an individual who is equipped with the right experience and skills required for this role.”

According to Ibrahim, each agent goes through a 12-week training course specifically focused on the cultural nuances of a particular market.

Yasmine Aldamiry, Uber MEA safety team lead, added that spatial real-time mapping tools, integrated communication software applications and customer relationship management tools play a crucial role in helping the agents to quickly and effectively respond to safety-related queries.

High-tech investigations are also necessary to flag fraudulent activities, such as false driver or rider accounts, account take-over and discerning incorrect or inaccurate information provided during a reported incident, she noted.

Aldamiry referenced a recently reported incident where a female rider accused a male Uber driver of raping her, shortly before dropping her off. The fist point of investigation was running an analysis on the profile of both the rider and driver from the time they first signed up for the app.

“We then used software that is able to playback the entire trip on a mapping system, tracking possible stops, the speed of the vehicle and any other abnormal activity. Using our system, we were able to piece together the entire journey, which showed there were no stops along the way and the time of drop-off, to the time of the next pickup, among other details.

“The investigation showed it was not possible that the driver had any spare time to rape the passenger. The lady later said she was intoxicated and was raped by somebody before falling asleep, but she had not remembered who the person was,” explained Aldamiry.

Share