ZineOne, a California-based startup founded in 2013 is all about transforming the way enterprises interact with their customers on digital channels such as web, mobile app, ATMs, kiosks, and in branches or stores. Bringing real time, contextual interactions to enterprises, ZineOne is enabling their clients to connect with their customers in a manner that ensures relevancy and interaction. To elaborate, its the relevancy in terms of customer’s immediate context and interaction based on real-time patterns and triggers.
“We are focused on continuous application and training of machine learning models on streaming data. We are helping enterprises implement an intelligent decisioning layer across an ever-growing list of digital channels”, explained Aurobindo Sarkar who heads the Engineering team in India, while interacting with Analytics India Magazine.
Idea behind its conceptualisation –
It’s an interesting story on how ZineOne came into existence. Sarkar said that the concept behind ZineOne solution is what many may call today ubiquitous computing. “It’s almost like what was conceptualised in the late ‘60s novel 2001: A Space Odyssey, where HAL 9000 was a sentient computer that could control your surroundings, communicate, interpret emotions, and recognize faces”.
Similarly, ZineOne can be thought to be the brain behind an enterprise’s brand-to-customer and intelligence-based interactions, exclaimed Sarkar. This brain can absorb everything from what a user is doing right now to what the user has done in the past, and seamlessly integrates it to understand the trends, patterns and intent.
Founded by Debjani Deb, Manish Malhotra and Arnab Mukherjee, ZineOne uses intelligence to engage with customers in the context of what is relevant. Let’s find out how!
Analytics and Machine learning at ZineOne –
ZineOne says that it is like a brain that carries context for user at every stage of their journey across digital channels. But how exactly it does that? Sarkar is quick to add that analytics and machine learning are critical here.
“Today, an enterprise is inundated with a high volume of customer data from a variety of digital sources, and in most cases, this data is stored, processed, and analysed to reach suitable decision. The inability to take advantage of data in motion to take immediate decisions may lead to significant loss of opportunity and revenue”, he said.
This is where machine learning comes into play, as ZineOne enable enterprises to not only gain insights from user activity on their digital properties, but also enable actions in real time within the customer’s’ immediate and most relevant context.
“This real-time decisioning based on user activity streams to execute contextual, relevant and personalised actions; enhances user experience, boosts transaction success, and increases revenue for our customers”, shared Sarkar
ZineOne at work, ensuring consistent user experience –
The startup is redefining the brand-to-user interaction in real time by bringing power of streaming analytics to customer and machine interactions. Explaining the same, Sarkar said that suppose a user is browsing for auto loans on a bank’s website on her desktop, starts a loan application, but leaves the website without completing the application. In such a situation, ZineOne sends web push to this user, reminding her about the incomplete application.
He added, this web push will typically have a link that could take the user to the target page within the application exactly where they left off. If the customer ignores the web push, but next day opens up their mobile app to pay bills, ZineOne can still recognise this app user and show a message about the incomplete application, enabling them to continue where they had left.
In case the user chooses to dismiss the reminder about the incomplete task, the system will remember their preference across devices and platforms, and not show the reminder again.
Here, ZineOne is maintaining a persistent context for the customer, resulting in a seamless consistent user experience across channels.
Industry use cases –
The startup has partnership with HDFC bank, among others, which is benefitting more than 40 million customers of the bank. “We are looking forward to working with HDFC to enhance customer experience across their digital channels. Our out-of-the-box use-case enablement approach combined with dynamic events and fine-grained customer segmentation will help us execute HDFC Bank’s customer engagement initiatives within minutes or hours rather than days or weeks”, said Sarkar.
Along with plans of associating with other banks, they are helping their customers rethink their customer engagement designs to include real-time activity patterns and trends. “We are also providing deeper insights and analytics that are helping our customers improve their product designs”, he added.
Growth story –
On being asked about the growth story, Sarkar said that it has been exciting to introduce a transformative technology solution to the market. “As we help our customers enable increasingly creative and engaging use-cases, we feel our solution being validated on a daily basis”.
“As the product and its usage evolves in the coming weeks and months, we feel the momentum building up for more advanced analytics and AI / ML technologies driving key marketing and product design decisions in the enterprise”, said Sarkar.
The growth that ZineOne has witnessed is also evident from the funding that it has been receiving. The startup claims that Hyderabad Angels has been a solid supporter, along with Shri BVR Mohan Reddy, a Padma Shri awardee, joining their board. They look forward to use these funds to further strengthen their product offering and rapidly evolving their streaming analytics and machine intelligence related features. “Our target is to be recognized as a leader in the cloud-based stream processing market”, said Sarkar.
Concluding note –
Sarkar said that bringing AI and ML technologies in the space of real time, contextual interactions require customers to understand big data technology stacks, cloud technologies and the associated infrastructure-related investments. “There is a significant hype in these areas, however one thing is absolutely clear in our minds that there are very few who can do real-time, contextual, and intelligence, really well, at this scale”, he said in the closing.
Try deep learning using MATLAB