2016 is shaping up to be the year of bots. It seems obvious judging by the incredible amount of articles written on the subject and their popularity with brands. New bots are created every day for messaging apps.
But how can this trend become useful for brands? Can bots become a tool for a successful customer service strategy?
How can brands distinguish the difference between buzz and real opportunities to transform their customer experience? Can artificial intelligence make bots increasingly intelligent? Should our human customer service agents fear the loss of their job in favour of bots?
This article aims to help you better understand chatbots and learn how iAdvize can meet your expectations on the subject.
What’s the context?
Chatbots are computer programs with the goal of automating conversations. Nothing new there, bots are not an innovation as such. In the 1960s, MIT researchers dreamt of initiating conversations between machines and humans with the help of the Eliza computer program. More recently, virtual agents appeared on websites to handle basic questions and transfer the conversation on to human agents when needed. Most of the time, this process proves to be quite efficient. iAdvize has been working with market players offering virtual assistants such as Virtuoz, Living Actor and Askom.
For their part, some digital market players have already developed their personal assistant (Apple Siri, Google Now, Facebook M, Cortana, etc.). These programs can interpret oral requests and respond to them. They introduced users to this type of interactions by directing research requests towards artificial intelligence and language processing.
But now, a new factor needs to be taken into account: half of the global population connects to a messaging app at least once a month. Everybody now has one or more of this type of apps installed on their phones, which results in them abandoning brand apps. As a result, brands are now thinking about new ways to reach their audience via new touch points: the messaging apps. Users see messaging apps as an easy and quick way to communicate with brands. It’s very likely they will be using messaging apps just like they use social networks to interact with brands.
Messaging platforms needed to adapt their customer engagement strategy and let developers interact with their technology to develop their bots. Kik and Telegram did exactly that right before Facebook announced the launch of its “Messenger Platform” during this year’s annual F8 conference (annual global developer conference).
Since these announcements, the craze hasn’t stopped: it is fueled by opportunities created by available technologies that are increasingly advancing. APIs facilitate the communication between apps, machine learning and language processing and offer new perspectives, especially since Facebook and Google make available their algorithms and the results of their research on personal assistants (Wit.ai for Facebook and the Cloud Natural Language API for Google).
In this context, new startups are flourishing on the Bots On Demand or BaaS (Bot as a service) markets. It enables everyone to create his or her personal bot and to diffuse it on one or more platforms.
Thousands of bots now exist, and hundreds of articles have been published to comment on them (and usually criticise them). Originally developed in the entertainment and service sectors (the CNN bot was launched on Messenger right after the F8 in April 2016), we are now expecting a lot from bots regarding customer service.
Chatbots & Customer Service
Automating customer service has always been the greatest fantasy of most brands. But they will need to wait a few more years for their dream to come true. Bots are not yet able to show compassion. They are not yet able to think out of the box to provide a personalised solution to a particular problem. Users still want human interactions; they want to communicate with their peers. And every day, even if answers are available on websites via FAQ pages or Web-Self-Service, many users are still asking agents fundamental questions via live chat or over the phone monopolising contact centres.
At iAdvize, we think that combining automation and human expertise can create a remarkable customer experience. Bots can help humans but cannot replace them. On a brand’s website, on its mobile app or a messaging app, bots will be able to take over the most fundamental questions from visitors and carry out the easiest repetitive tasks to help them. Bots will then pass on conversations to human agents when needed. This way, agents will be able to focus on questions with high added value and offer their expertise to visitors who most need their help.
To offer this type of experience, brands will encounter several significant challenges:
- Identify the right scenarios
When should bots intervene in a conversation? Brands have to know why they want to use bots and think about the mission and objectives they want to give them. Which questions should bots answer? What type of conversations should the computer program learn? Which rules will enable the bot to handle a conversation towards an action? What tools should the chatbot use?
- Capitalise on existing knowledge
BaaS tools don’t yet enable brands to integrate their existing knowledge base or their data for customer service with chatbots. Duplicating all this knowledge in a BaaS solution takes a lot of time and effort and too much maintenance in the long run.
- Define situations when a human agent will take over and how
Some BaaS tools offer the possibility to transfer conversations from bots to human agents, but it’s usually only possible within the Messenger platform. Brands should then define at what stage of the conversation or from which actions a human agent should automatically take over.
- Offer the possibility across platforms
Visitors now use several touch points to interact with brands: the website of a brand, SMS, Messenger, WhatsApp, Twitter, Facebook, etc. A bot should be able to handle conversations from all of these touch points and enable human agents to take over via a tool which centralises all requests coming from them.
How the iAdvize platform addresses these challenges
Ever since 2010, iAdvize has been a major market player in the online customer service sector. iAdvize knows the topics usually brought up by visitors and how conversations are conducted thanks to data compiled from conversations on the platform. This data is a fabulous asset to help brands determine which subjects bots should learn to respond to and how they should interact with visitors.
By capitalising on existing best practices, one can quickly identify which data is necessary to qualify a contact and when human agents should take over bots in a conversation.
Furthermore, the iAdvize platform already helps customer service agents to formulate their speech by offering knowledge base solutions (library of canned responses, links and offers). Thanks to Smart Answers (a solution currently in beta test on our platform), we can give agents suggestions they can use to answer their visitors more quickly. These recommendations are based on past discussions they’ve had with visitors. All these tools enable us to build the bot workflow and give it a sustainable knowledge base easily.
Finally, iAdvize is a conversational commerce platform. From one single tool, we enable agents to initiate conversations via online touch points, SMS, Facebook, Messenger, Twitter, Instagram and soon from other new contact channels.
Eventually, the integration of a bot will involve the integration of a new agent and a multitude of new possibilities: from the optimised addressing of conversations via our behavioural targeting engine to the detailed analysis of their performance via the real-time reporting of KPIs.
Are you looking to integrate bots in your customer care strategy? Request a discussion with one of our experts.