CapitaLand Investment enhances operations with data-driven AI models built using Azure OpenAI Service

“CapitaLand uses digital technology to drive business outcomes and deliver greater value to our customers and partners,” shares Hwee Lim Wong, Head of Digital International at CLI.

Bringing data together

CLI’s journey toward a data-centric culture started in 2016. Each of its business units used to have its own on-premises data warehouse. To bridge and consolidate the company’s data, CLI merged its data warehouses into a single data platform on Azure. "Before, our employees had to use multiple reporting tools. Moving to the cloud allowed us to consolidate data and standardise our tools and software," explains Wong.

This shift significantly improved CLI’s operational efficiency and data visibility. "Today, we have more than 4,000 users from more than 40 countries consuming data from this platform 24/7, something we could never achieve with disparate platforms," notes Wong. “Plus, now that we use one solution instead of many, we have simplified internal skill sets and adoption, both for our internal and external teams. This enables us to offer self-service training more efficiently.”

Using AI to accomplish more with less effort

Having established a unified data platform, CLI could now accelerate the adoption of AI technologies in its processes. For example, its investment team wanted to have a more efficient way of identifying and analysing data, moving away from manually combing through extensive economic and property data and using Excel sheets to unearth more insights. “Since we have consolidated data on Azure, we realised we could run predictive AI models and assist our investment teams in identifying suitable markets or cities for investment,” recalls Poh Har Neo (Dr), Vice President, Head of Data and AI at CLI.

The company developed AI models using Azure Cognitive Services for its investment teams to explore and analyse deals at scale and extract key information.

CLI also created an enterprise chatbot to handle a range of internal functions, from drafting investment papers to creating marketing communications, all while ensuring data security. “We’ve used the chatbot for emails or speech writing in English and Mandarin. Though chatbots aren’t new to us, this generation is much more accurate and has generated lots of positive feedback from our team,” says Wong.

Achieving simplicity and accuracy with AI

Besides making processes more efficient internally, CLI also sought to improve customer experience with Cubby, a unique hospitality operator chatbot for Ascott. Ascott is a trusted hospitality company with about 950 properties globally, which offers serviced residences, hotels, and more. Built with Azure OpenAI, Cubby uses real-time data from Bing search alongside information available on Ascott property pages. It assists guests in planning itineraries and answering inquiries. In the future, it will also assist customers with their reservations.

Ascott has implemented an AI model internally to optimise room configurations within each property, a critical aspect of the company’s development and enhancement process. This initiative aims to tailor accommodations to traveller demand, maximise revenue for property owners, and efficiently manage apartment inventory. “The AI model will consider different factors like past bookings and competitors’ offerings to suggest new configurations,” explains Wong. “For instance, a two-bedroom apartment for a two-person booking is not necessary when a one-bedroom apartment would suffice. This helps to save space and increase our overall apartment inventory.”

CLI’s retail arm has also incorporated AI into its rewards program, CapitaStar. Customers can earn reward points shopping at CLI’s malls, scanning their physical receipts using the CapitaStar mobile app. "Previously, a customer would have to wait three to seven days for their receipt to be processed and receive their reward points,” remarks Dr. Neo. “With Azure OpenAI Service and optical character recognition technology, their receipts are processed almost instantly, and points are awarded within seconds instead of three to seven days."

More insights at a lesser expense

With a unified data platform and AI technologies, CLI can now do more with less. "Moving to Azure has enabled us to save more than SGD1 million in operational costs while being capable of handling increased data volumes,” highlights Wong. The investment team can now evaluate more deals and swiftly identify suitable investment opportunities to deliver value to stakeholders. The company's internal team efficiency has also improved substantially.

Fostering innovation and growth

“We looked at many competitive platforms when choosing the technology we needed. What kept us coming back to Microsoft was the support we received,” shares Wong. “From day one, they’ve been with us every step of the way, providing solutions and training for user adoption. The data platform can also grow alongside CLI as we continue to scale our business.”

Looking ahead, CLI is focused on further cultivating a culture of innovation and expertise in AI among its employees. “This year, we conducted a data and AI symposium with nearly 1,000 employees participating,” Wong shares. The event, which also featured external guest speakers from Microsoft, PwC, and Accenture, aimed to spark inspiration for employees to continue exploring AI and propose new use cases. “Our goal is to encourage employees to gain hands-on experience and get comfortable with the idea of citizen development, where most of our employees could develop solutions for themselves and our customers,” concludes Wong.