User Experience (UX) in IT is typically thought of as a combination of intuitive UI coupled with robust user engagement workflows. In the world of chatbots, however, a strong UX is increasingly associated how a bot uses language to engage with a customer – what could be called the bot’s “personality”.
As Natural Language Processing (NLP) and Artificial Intelligence (AI) improve, the ability to add a semblance of charisma to a chatbot is becoming more of a reality. Natural Language Processing is the technology that interfaces between user input (language) and system output.
Behind the scenes, NLP technology associates user input with an “intent,” which is then correlated with an appropriate output. The manner in which that output is conveyed is the province of a bot’s personality — effectively the chatbot’s UX.
From a user’s perspective, the technology under the hood takes a backseat to the direct experience of chatting with a bot. When configured correctly, a bot’s personality aligns with the organization’s brand and is consistent with the chat’s purpose, and thereby helps the user have a positive engagement with the company.
Customized Chatbot Personalities
Let’s look at how some different chatbot personalities can be matched with different chat scenarios:
- Banking Chatbot: Generally-speaking, financial institutions value a culture of trust, responsibility, and reliability. This means a chatbot that is professional and polite during all interactions, with a focus on being succinct — particularly when performing banking transactions. If the bank’s marketing brand is more informal, then there is still plenty of room to convey friendly helpfulness during a chat… such as retrieving customer analytics and providing unexpected content (e.g., “You saved 30% more money this quarter than last! At this rate, you’ll hit your savings goal in 45 days. Great job!”);
- IT Help Desk Chatbot: In this sector, more often than not the customer will enter a new conversation in a compromised mood. Customers want an answer quickly and are likely frustrated. The last thing an angry customer wants to deal with is an overly verbose chatbot that leads the customer in circles without providing a resolution. In this all-too-common scenario, an effective chatbot needs to combine a sense of empathy with a strong focus on issue resolution. Advanced NLP/AI may be able to determine the level of frustration based on the user input and then limit the resolution process accordingly in favor of escalation to a human agent. This capability introduces a sense of chatbot customer awareness that, over time, would improve when Machine Learning technology is applied; and,
- Travel Agent Chatbot: This type of chatbot excels at learning about user interests and then matching those interests with appropriate travel products. In this scenario, a chatbot’s personality should be helpful and informative with room for light-hearted interactions. In this type of upsell-focused industry, chatbots with an appropriate personality can suggest travel packages and use modern marketing techniques to increase overall sales. The complexity of these types of transactions often result in businesses using a hybrid approach whereby a chatbot starts the sales process and a human sales agent completes the interaction.
Overcoming the Limitations of Chatbots
As advanced as AI, Machine Learning, and NLP have become, the current reality is that chatbot technology can still be problematic. Given the complexity of human language and the nuances of communication, this should be expected.
Current chatbots using advanced NLP can also be resource intensive, particularly when big data Machine Learning comes into play. By purposely focusing the scope of chatbot functionality to specific purposes, developers are given more freedom to dedicate resources towards bot personality configuration.
Some common techniques to address these challenges include:
Open-ended chats where the user can discuss anything under the sun with a chatbot may be enjoyable, but in reality it is not effrective. In nearly all chatbot implementations, the scope of conversation is pre-defined and communicated before a chat even begins.
For example, a complaints management chatbot may establish chat boundaries by saying, “Hi! I’m here to guide you through the complaints process. I can help you to lodge a complaint, check the status of an existing complaint, or provide information about our resolution process. What would you like to do?”
By limiting the conversational scope, chatbots are more likely to provide a helpful experience and avoid the pitfall of providing irrelevant answers or simply not understanding customer questions.
Talking with a personality-infused chatbot does require a suspension of disbelief — developers hope that, over time, that disbelief will diminish as technology improves. Chatbot personality designers have worked to mitigate this disbelief by being upfront with the reality of chatbot interactions via self-deprecating humor. It’s not uncommon for chatbots to blatantly state, “I’m sorry that I’m boring you with this list of options, but what can I do? I’m just a pre-programmed robot!”
By providing users with the proverbial “wink-to-the-camera” and breaking the fourth wall, chatbot developers are acknowledging current technological limitations while asking you to cordially play-along.
Conclusion – Consumers Value Bots With Personality
When the concept of Artificial Intelligence first appeared, society seemed to coalesce around the sci-fi notion of an “ideal robot.” Whether it’s HAL from Space Odyssey or androids from Blade Runner, robotic personalities always seemed to be dutiful, obedient, and wholly acquiescent to user demands.
In the world of chatbots, organizations are discovering that customers value chatbot interactions more when the bot is relatable and emulates human emotions. While the technology is still in its early stages, the fact is that corporations are currently enjoying increased revenue due to the use of chatbots in their customer experience workflows. The quintessential 1-dimension robot of sci-fi is rapidly being replaced by bots that ask about your day and make small-talk. Businesses expect that customers will buy into the fantasy and willingly forget the reality of their robotic chatmate’s true nature… and, so far, the financial returns are proving them right.
Crow Canyon Software delivers the power of a connected, digital workplace to your organization, helping you be more successful and competitive. We provide the tools that allow you to gain the maximum benefit from SharePoint and Office 365, Microsoft’s premier collaboration platforms.
Want to learn about how our business productivity applications for SharePoint and Office 365 can power up your organization? Give us a call at 1-925-478-3110 or contact us by e-mail at firstname.lastname@example.org. We look forward to hearing from you!