A Multichannel Chatbot: 5 Questions to Ask Yourself Before You Get Started

Today’s consumers are no longer satisfied with just one channel. Their shopping experience is punctuated by numerous points of contact—including websites, social media, brick and mortar stores and marketplaces—and a variety of devices—including computers, smartphones and tablets. This is the very definition of “multichannel,” a reality that inevitably affects the implementation of a chatbot. Here are five questions to ask yourself if you truly hope to transform your bot into a multichannel conversational agent.

1. What exactly is a multichannel chatbot?

When we use the word “multichannel”, we’re referring to a customer experience that multiplies the points of contact, an experience that will vary according to the media consulted and the device used. However, in the context of a chatbot project, the feed of the channel, whether textual, audio or visual, must also be taken into consideration. If multiple screens make choosing the ideal channel more complicated, it’s possible to imagine a bot that can interact with users through different channels.

2. What channels are best suited to your users?

In order to be efficient, your multichannel chatbot must favor the channels that are used by your clients. There’s no point in implementing a chatbot on Messenger if only five percent of your prospects are on Facebook. On the other hand, a service company with employees who are often in transit would benefit immensely from integrating a mobile channel into its bot.
Therefore, you should start by determining which channels you really need: what is the typical user experience in the case of your clients? At what point do they require assistance? Through what channel do they typically interact with your chatbot? To make the right decisions and ensure that your conversational agent evolves in the right direction, you must be well-acquainted with your personas.

3. How do you provide value added service in a multichannel format?

The members of your target audience are all different from each other. And so are your needs in terms of channels. In order for your chatbot to provide genuine value added multichannel service, it must answer the same questions and furnish the same information, but across different channels, in accordance with the habits of your users. And it must always include the capability of generating a help desk ticket in the event that it’s unable to respond to an inquiry!

4. Is it possible to adapt the ergonomics of the chosen channel?

According to the channel being used, there’s the matter of ergonomics. Take a voice chatbot, for example: it’s a practical option for responding to users wishing to obtain information quickly without the hassles of written text. The method of reproducing the spoken word differs greatly from that of the written word, with different constraints in terms of personalization and configuration—choosing a voice, a rate of speech, a tone, inflections, etc. This represents a genuine challenge for your multichannel chatbot. Especially if you need to offer multiple methods of reproduction, adapted to multiple channels and audiences…

5. How do you ensure a consistent customer experience across multiple channels?

Your users aren’t focused on just one channel. In most cases, they go from one channel to another, from their desktop to their mobile devices, from a website to social media and from text to images. So the challenge is to provide continuity, without losing the data that has been collected through each channel. Your chatbot must be able to record and recover everything the user says and writes!

See also: Is AI indispensable for your chatbot?

The prospect of a multichannel chatbot raises numerous questions, especially with regard to its interoperability across channels. The most important thing is for your users to avoid having to start from scratch each time they access the bot through a different channel, through a different device or using a different type of feed. Do you need advice as to which channel (or channels) would be most appropriate for your chatbot?

Photo credit: Fotolia / Bulat