To cater to the needs of data hungry and digitally savvy customers, Starhub launched Giga, a fully digital mobile service with its own technology platforms and customer support. To increase awareness and usage of the Giga mobile app, the team prioritised the app’s user experience and created an AI chatbot so powerful that they clinched the award for eCommerce Brand of the Year 2021 and took a finalist position for Best in eCommerce (Brands) – Electronics & Gadgets at MARKETING-INTERACTIVE ’s Asia eCommerce Awards 2021.
Giga faced two key challenges within Singapore’s telco sector. Firstly, there was intense competition with more players joining the Mobile Virtual Network Operators in Singapore. The local telco scene had more than 10 major players serving a small but price sensitive population with a mobile penetration rate of 152%.
Secondly, Giga needed to deal with evolving customer demands. Customers wanted value plans without being tied to a mobile contract laden with terms and conditions. They also wanted high-quality, reliable and seamless online experiences for secure transactions. There was also an emphasis on customised experiences, convenient payment methods and the need for operators to be responsive to queries.
With these challenges in mind, Giga’s mobile app was designed with the following benefits:
- Real time service mobile platform which enables customers to purchase booster packs immediately along with real time usage monitoring and alert triggers
- Scalable app and web platform that allows for seamless customer journeys across both platforms
- Frictionless payment flows to facilitate digitalised payments
- Digital ID verification to detect fraudulent transactions
- Partnership with delivery firm that offers next day delivery for orders submitted before 6pm
- Fully digital AI powered chatbot named Gino to serve customers on Facebook Messenger, Whatsapp, Giga mobile app and Giga website
The most innovative aspect of the app was the creation of GINO, an interactive chatbot with the ability to engage in intelligent conversations on channels like Facebook, Whatsapp, Giga app and Giga web. GINO is able to answer predictable questions and can offer small talk enablement with localised language such as “Singlish”. GINO is trained regularly on multiple aspects to enhance customer experience through continuous machine learning. This omni channel support powered by GINO and live chat addresses the lack of customer support within the industry. Giga’s strategy involved visualising the end state where the team started with the customer experience and worked backwards to the technology.
With a growing subscriber base, the team saw a need to enhance its AI chatbot GINO to understand what customers were trying to ask even before they typed it out. This involved NLP and Deep Learning (in the form of Recurrent Neural Networks) to power conversations. Auto suggestions were prompted to customers in the form of guide buttons to speed up the ticket resolution process.
To enhance GINO, the following techniques were used:
- Syntax Analysis – Ability to derive the meaning of the conversation from unstructured, grammatical incorrect sentences or misspelt words.
- Synonym Analogy – Ability to understand the same intents with different but equivalent words
- Sentiment Detection – Ability to understand sentiments, emotions and tones expressed through text conversation and respond in an appropriate manner to the conversation accordingly.
- Contextual Recognition – Ability to remember and correlate the context of its interaction and respond appropriately.
- Omission of Sensitive Data – Ability to filter sensitive information before sending the data out to the internet cloud-based NLP for processing.
- Friendly Message – Ability to respond with a friendly message if the answers to the intents are not found in the database.
- Machine Learning – Ability to have the means to self-learn through users’ conversation and increase in language understanding.
GINO appears on the top right corner of every page on the mobile app and webpage, ready to lend support anytime.
Characteristics of GINO:
- Conversational Maturity – GINO has specific natural language processing (NLP) capabilities to understand the context of a conversation in multiple languages. GINO can also identify the intent of a question–what is needed– to provide an accurate first response, and propose options to confirm or clarify intent.
- Omni-capable - GINO converses seamlessly across multiple digital channels and retains data and context for a seamless experience. GINO serves non-Giga and Giga customers on Facebook, Whatsapp, web and app
- Emotionally Intelligent - GINO can infer customer personality traits and understand sentiment and tone during an interaction to deliver a personalised experience or escalate to a live-agent when necessary.
- Free to Explore - GINO can reach, consume, and process vast amounts of data– both structured and unstructured–to surface insights from any source - to gather relevant data to solve customer issues quickly.
- Pre-Trained - GINO is pre-trained to understand brand-specific or industry-specific knowledge and terms. GINO is also pre-configured to resolve common customer requests of the industry like how to port-in, how to sign up, troubleshooting network problem or an APN setting.
The continuous focus on improving the mobile app features and use it organically to drive upsell and referrals allowed the team to create a product and user journey that customers truly wanted and enjoyed using. Giga’s customer first mindset also helped to create brand trust where customers willingly spread the word to promote Giga to their friends and family. This ultimately led to significant growth in subscriber base and active users.