Doing business in an old fashioned way… Talking!

Written by: Vinicius Mesquita

February 13, 2017 - Exploring conversational commerce

Since January 2015, when Chris Messina coined the term ‘Conversational Commerce’ for the very first time, we all realized there was something smelling good in the air. Many tech giants (Apple, Google, Facebook, Microsoft, Amazon) have already made significant investments in their voice assistants or have shaped their messaging apps (Messenger, WhatsApp, Allo, WeChat) to run as independent business platforms, allowing them to do business in the most traditional and natural way: through conversation.

For those who are not familiar with the term ‘Conversational Commerce’, it means that consumers can focus on their needs (shopping, banking, travel booking, healthcare) and interact through voice, messaging, and chat apps with a company that provides the services or goods they are interested.

From the end of 2015, messaging apps surpassed social networks in terms of use and became undoubtedly the ‘new way of talking’, opening space and consolidating the concept of Conversational Commerce which one year earlier could be perceived as just a smell. According to Gartner, by the end of 2016, more than $2 billion in online shopping will be made exclusively by mobile digital assistants.

A pinch of AI

When we are talking about this ‘new’ way to make business, what’s even remarkable is that consumers are willing to use the communication channels they already enjoy in combination with advances in Artificial Intelligence, Machine Learning and Natural Language Processing. Once combined properly, while still maintaining the peer-to-peer interactions that consumers seek, the results create endless business opportunities.

On top of these recent advances, this channel benefits of two aspects typically provided by chat apps, which enhance the experience into Conversational Commerce: First, the use of asynchronous/synchronous communications adds convenience by allowing consumers to initiate a chat and end later from the same point where they stop it. Second, making the chat itself as a source of context and historical data, avoiding consumers to repeat everything when a change in the touchpoint occurs.

It's happening now

In October 2016 during the Money 20/20 event in Las Vegas, at least two big players showed their last developments in this field. Bank of America unveiled ‘Erica’, an artificial intelligence text and voice bot capable of analyzing a customer’s financial data and proactively helping with payments, savings, and common queries. According to the head of digital banking, Michelle Moore, they hope to help consumers create better money habits.

Mastercard also presented a new chatbot to be launched in the U.S. in 2017 called ‘Mastercard KAI”. It aims to fulfill customer requests and solve problems, enabling financial institutions to create entirely new consumer experiences to help them deliver better mobile experiences using Facebook Messenger. They are working together with Kasisto, a company behind the conversational AI platform KAI Banking, and used already by the Royal Bank of Canada and the DBS Bank in Singapore.

The KLM Royal Dutch Airlines is one of the companies that can be already proud of the usage of this kind of platform. They launched their services through Facebook Messenger and WeChat last year and it’s possible to access all your flight documentation like your book confirmation, itinerary, boarding pass, check-in confirmation and even delay notifications or reminders when check-in opens. To offer even more convenience to their customers, they allow passengers to change their seats using this chat channel.

There are a lot of different companies such as eBay, H&M, Domino’s, 1-800-flowers and Sephora that already use this technology to generate more business and provide a better customer experience. This is not a surprise, considering a Facebook study which revealed that 53% of consumers are more likely to shop with a business that they can message directly.

Time to explore

Until now, these companies seem to be doing well, although not everything is a bed of roses. Despite the buzzword, media trend, and consumer hype, the technologies behind chat bots are still evolving. At this point, they are still unable to provide natural and fluid conversations.

Savvy consumers expecting bots behaving like humans, with full access to their transaction historic, preferences, and full profile, might be disappointed with the output from this current phase. At the same time, early adopters can benefit from these services and companies can learn and evolve their AI engines behind the scene.

Currently, it makes sense apply bots in industries with a wider tolerance for failure or, at least, in services that cannot serious compromise a customer. It’s quite reasonable that retailers take the forefront in this offering and financial services and healthcare wait a bit longer to provide experiences in the same level.

5 Key elements for success

Besides the whole care that companies must have with contextual conversations to offer a seamless customer experience, provide services through chat and message apps implies in all kind of challenges and it should not be different of do business through any other channel. Perhaps your bot can not be blamed if something wrong happens, but your company will be liable likewise.

  1. Add value
    It might sound pretty obvious, but prior to get into the development and technical decisions, it’s crucial define a strategy based on what service your company want to offer and what value it’s being added from a customer perspective.

  2. Security
    As any other digital channel, it’s fundamental consider which type of service is being provided and ensure that the authentication method and equivalent level of assurance are properly applied in order to keep either customers and companies safe.

  3. Privacy
    Transparency is certainly one of the key points to keep trust in the digital world and a primary aspect when it comes to data property and privacy. Establish from the beginning what kind of data will be used not only for the purpose of the primary output of the conversation, but what and how personal data will be used to improve this channel and the quality of a chatbot assistant.

  4. Hybrid approach
    Despite what companies can do to train and improve their bots, real situations usually deal with unusual requests and might be a big challenge even for an advanced and experienced system. Many successful bots are using a sort of ‘Hybrid approach’ as a way to handle some limitations on natural language recognition, providing a way out to a live chat or even a call with a human being.

  5. Interoperability
    Highly expected from a digital channel, integration with other channels and access to information based on legacy systems must be consider and added to the equation. Set expectations of what consumers can do and be sure that other channels can, at least, reach information from chatbot assistants.

So in the same way your company should not simple go mobile without a clear strategy, the fact that now it is possible do business through chat, 24 x 7 and using Artificial Intelligence to make bots more reliable should not be enough (even if it sounds sexy to you). Focus on the customer experience, plan how to deploy a solution considering all dimensions involved, consider secure by design, test all together, put into the oven and wait for the good smell that we talked in the beginning.

Disclaimer

These are the personal opinions of UL’s employees and its guests and should not be misunderstood as representing the opinion of UL's clients, suppliers or other relations.