MENU

Omni-Channel Customer Engagement Article

The Right Way to Build a Chatbot

October 06, 2016



Do you like long, flowery conversations with a robot? Television has given us plenty of opportunities to view these kinds of interactions via Sci-Fi and futuristic shows and movies, yet the thought of actually engaging a chatbot doesn’t really hit the same level. Yet for a number of companies trying to create the omni-channel customer engagement experience, this seems to be overlooked. 



A recent blog post by omni-channel customer engagement solutions provider, Aspect (News - Alert) focuses on the chatty customer service chatbox and why it’s a bad idea. Those who are developing a chatbox may assume that giving it a personality is one of the ways you can enhance the customer experience. But the focus should be on delivering results that make customers happy, not attempting to build a relationship.

When building any type of technology to enhance the omni-channel customer engagement experience, you have to understand why and where customers are contacting you, as well as their overall goals. We’re accustomed to personality in Siri or Cortana as these technologies are developed as artificial intelligence (AI). By contrast, customer support chatbots are a self-serve medium where the focus is on natural language and machine learning.

While chatbots are considered the optimal, streamlined versions of live customer service representatives, there is a significant amount of training and resources used to help live agents best engage with customers. The chatbot with a lot of personality that is supposed to be helping a customer check a bank balance or the status of a mortgage payment could end up being annoying and actually distract from the end goal – not something that generally produces a happy customer.

The way the chatbot interacts with the client is also heavily reliant on the channel used to contact the company. If the customer is using text or Twitter (News - Alert), for instance, character space is at a premium and messages need to be concise and to the point, often even using acceptable shorthand or symbols. Customers are using these channels because they want short and quick interactions. If the chatbot were instead used on a voice channel, giving it a little personality so the interaction feels more human and is easily understood is a good thing.

Ultimately, any technology put in place to support omni-channel customer engagement has to do more than just consider the costs and complexities of the typical customer interaction. You also have to consider end goals and optimal outcomes demanded by the customer base that lead to satisfaction. Without this as your primary target, you’re not delivering on the experience.




Edited by Maurice Nagle
Article comments powered by Disqus