At TechSparks 2021, Tata CLiQ's Kamal Kumar and Vineet Mahale decode how data, behavioural science can elevate CX
As COVID-19 pandemic changes the way people spend their money and interact with brands, personalisation of user journeys has proved to be an effective tool for businesses to gain a competitive edge.
With the increase in available data, it is no longer uncommon to see companies expanding their data and analytic capabilities to derive actionable insights and connect more closely with their customers. However, amid the data-savvy culture, one often wonders about the accuracy of such plug and play solutions that enable businesses to send personalised notifications on the basis of customer data.
Kamal Kumar, Chief Data Officer, Tata CLiQ
During an insightful masterclass at TechSparks 2021 — YourStory’s flagship technology, innovation and leadership summit — Kamal Kumar, Chief Data Officer,and Vineet Mahale, Business Insights Manager, Tata CLiQ deep-dived into how Tata Unistore's Tata CLiQ and tried to solve for
erratic customer interactions by combining behavioural economics and data science.
Tata Unistore operates three e-commerce stores — Tata CLiQ, Tata CLiQ Luxury and soon to be launched CLiQ beauty app. Since its launch in 2017, Tata Unistore's platforms have received 900 million+ visits and 117 percent year on year growth in orders.
Here are the key takeaways from the session:
Need for more nuanced insights for customer relationship management
Most of the plug and play available in the market capture data about consumer behaviour and help you design app notifications based on those insights.
"While this was one way of customer relationship management (CRM) notifications, we were looking for a more nuanced way of deep-diving into how a user interacts with different categories. Is it customer first strategy to send a cart abandonment notification or whether customers feel iffy about it. Or if there is a section of customers who don't like getting notification, among other insights," said Kamal.
"In our experience, if a brand consistently uses these notifications to interact with users, they might end up either turning off the notifications or uninstalling the app. Hence, there needs to be a balance when these strategies are leveraged. Around 2-3 years back, when we were analysing our data, we noticed that customers are like to the point messages and highly contextual and personalised solutions," he added.
Overuse of AI/ML = Spamming customers' phones
Kamal believes that as the penetration of technology is increasing, so is the extent of spamming. "Technology is abused to send dozens of notifications, giving strong hand to clickbaits. We have been abusing the users' attention with artificial intelligence and machine learning-based messaging. These notifications might claim to be hyperpersonalised, but end up being deceptive,” he said.
According to Kamal, the company aimed at building a core platform that could power user communication and depersonalisation across different channels like CRM push notifications, storefront experiences or contact centres.
"We invested in creating a data lake and streaming data capabilities early on. We ensured that all data about consumer interactions came to the analytics stack and our data scientists have access to the most recent data at the granular level,” Kamal said.
“We ensured a culture of data-savviness and brought about transparency by making it super easy for business stakeholders to look up performance,” he added.
Not all data and insights are useful
Kamal talked extensively about how Tata Unistore laid down certain basic tenets for leveraging analytics so that the strategy isn't people-dependent. "For instance, we accept that data and analytics are not always neutral. We told our business stakeholders that even if some uses of data are profitable, but not good for our customers, then we will shy away from using it in an algorithm or model," he said.
He explained the tenet with the example of data on customer attributes like gender and age. "We generally don't use the information to train our models as it may lead to biases."
How Tata Unistore leverages behavioural economics
Meanwhile, Vineet delved into how Tata Unistore decoded customer journeys using behavioural economics and leveraging users' preferences to derive relevant recommendations depending on the phase of journey.
He chose the example of a male customer who's about to get married and is looking for ethnic wear to elucidate a point. Depending on his choice and scope of information, he browses different categories across price ranges and brands on the app.
"We analyse the sequential sessions to break down the user journey and its different phases. We derive biases in terms of rationality, choice overload, and information avoidance among others. These biases help us build a matrix to build recommendations for different sets of users. These recommendations are based on information like user interactions and browsing preferences, and are then sent as push notifications," Vineet explained.
In the event, the user makes a purchase, he/she is suggested other relevant categories for perusal.
To derive insights, Tata Unistore studies customer journeys at a very foundational level. "We study data on how users discover products, how is the journey different for users that have predetermined choices versus those who are explorers, the time usually taken between browsing and transacting, how this time varies across price points and much more," stated Vineet.
To log in to our virtual events platform and experience TechSparks 2021 with thousands of other startup-tech enthusiasts from around the world, join here. Don't forget to tag #TechSparks2021 when you share your experience, learnings and favourite moments from TechSparks 2021.
For a line-up of all the action-packed sessions at YourStory's flagship startup-tech conference, check out TechSparks 2021 website.