In the US, Amazon recently took artificial intelligence (AI) to a new level with the official unveiling of its offline store Amazon Go in Seattle, wherein AI algorithms can watch the various video feeds and identify who is picking up which item from the shelf.
In the US, Amazon recently took artificial intelligence (AI) to a new level with the official unveiling of its offline store Amazon Go in Seattle, wherein AI algorithms can watch the various video feeds and identify who is picking up which item from the shelf. The store works on the mechanism of a customer scanning a code on the Amazon Go app to enter the store and can pick items off the shelves. As he/she walks out, the items are automatically billed onto his Amazon account, thereby eliminating the need for cashiers or self-checkout kiosks.
Although not as developed, things on this side of the globe are catching up. In India too, several brands are waking up to the immense potential of AI and its effectiveness in reaching out to consumers. Take Myntra, which has designed its fashion line using its AI platform — Rapid. It picks up signals from the sales catalogue and public images from the web through geolocation to figure out what is trending. Currently, it has designed t-shirts, denims and bomber jackets for brands like Moda Rapido and Here & Now. The e-commerce brand is also designing an ethnic wear range using Rapid.
In fact, the Union Budget 2018 recently gave a major thrust to the Digital India programme, allocating Rs 3,073 crore towards AI, machine learning and IoT (Internet of Things). But as marketers battle with the challenges of automation, are consumers tickled enough about this new-age touchpoint?
The experimental stage
Sectors like health, manufacturing, BFSI, etc are using AI to quite an extent. “The retail sector is a little slow in using AI due to the presence of bricks-and-mortars, but is getting there through in-store innovations,” states Rakesh Barik, partner, leader — technology, Deloitte India. “In fact, the government is also looking at using AI to plan smart cities. But the jury is out on how it will change people’s buying behaviour.”
In fact, DDB Mudra, in partnership with Hotify, is providing AI-augmented products and solutions for precision consumer targeting, real-time prescriptive marketing insights, AI-augmented process automation, cognitive content, etc. Industry experts believe that we are in the early days of how AI can transform the marketing landscape. The most common usage of AI in India is chatbots which has yielded mixed results at best for marketers, according to Kabir Kochhar, Managing Partner & Business Head, The Glitch.While there are different approaches to AI such as NLP, semantic analysis, probabilistic modelling, etc, the most significant and profound impact is being created by machine learning/ deep learning, state experts. Chatbots are also increasingly being replaced by voice-assisted AI platforms that brands such as Ola, Zomato and HDFC are experimenting with.
One such platform is Ixigo.com’s travel assistant and recommendation agent (TARA). The brand’s vision is to make it seem like a human being when the user converses with ‘her’. “TARA is currently being fed data that we have collected over the years and will launch it in Q2 FY18. It is currently being used to handle customer support internally,” says Rajnish Kumar, co-founder and CTO, Ixigo.com. While the brand handles 60-70% of its queries without any human interaction, 60% of customer support queries are now being handled and solved by TARA.When Ixigo started looking at AI algorithms around four-five years ago, it just wanted to solve problems such as users complaining of wait-listed tickets and surety of confirmed tickets, which was easier to solve using AI and data unlike traditional algorithms.
“Interestingly, 30% of India’s searches about travel are voice driven and 80% are from tier II and III cities,” Kumar shares. Similarly, Myntra too is seeing an upside. For Moda Rapido, it saw a sell-through rate of three times with twice the margins for the AI-designed range as compared to the traditional lines. AI has been able to effectively help businesses decode patterns in their customers’ online purchase behaviour and predict the probability of a product return, resulting in cost optimisation. Most marketers use AI technologies for customer segmentation — leveraging customer data to create specific clusters of customers with shared attributes.
Jeyandran Venugopal, CTO, Myntra mentions that the brand is also using AI to automate its quality check process and make it act as a style assistant. It also plans to roll-out a post-order AI service that guesses why the user is calling, based on her order details, past calls, etc. “Voice/speech recognition is where we are investing in and are moving the platform to cater to tier II and III segments which are high growing,” he says. In fact, Rapid is also generating a lot of interest from other fashion brands, who want to license the technology. It helps the brand reduce the time taken for the selection of fabric and colours, helping it create a fashion line every 15 days, thereby ensuring efficiency.
“AI-driven fashion and technology will be a billion-dollar business in a couple of years. We are partnering with brands and are about to rollout some pilots for them,” Venugopal adds.
Milking the artificial cow
According to a recent IDC report, AI is expected to drive worldwide revenues from nearly $8 billion in 2016 to more than $47 billion in 2020, across a broad range of industries. “With the AI solutions by Boxx.ai, Nearbuy observed an overall increase in the click-to-open ratio by 35.7% in response to the email notifications that were communicated to the customers,” informs Prashant Mehta, VP and global service line lead — systems integration and data, SapientRazorfish. The point he makes is, when combined and designed with the consumer in mind, AI technologies can deliver solutions that drive customer loyalty, engagement, consumption and satisfaction relevant to stay ahead in today’s disruptive times.
Despite the scope, there are various risks and challenges associated with AI. The challenge for some brands lies in adoption, the time needed to understand the model and its efficiency. Brands are also worried about the customer experience, as chances of this going wrong are high. Nobody wants to create a bad experience deliberately; therefore, companies are worried about losing control to a machine. However, Kochhar believes the cost implications are not onerous in terms of heavy upfront expenditure and clients will be able to test the waters via AI solutions as added services through their existing vendors.
“Increased costs to marketers would be offset by improved ad performance, better RoI and richer consumer profiling. So I do see agencies pitching more AI solutions to clients in the near future.” The key is to identify the business problem that needs AI support rather than blindly adopting the technology. “Brands should also ensure that they work with a skilled solutions provider and also look at enhancing in-house team skills to maximise the potential of the solution,” summarises Deepak Nair, chief growth officer, DDB Mudra Group.