The COVID-19 pandemic triggered an immediate need for transformation of traditional business practices. Workers the world over were forced to work from home, and companies scrambled to redesign their way of working and business continuity strategies. One area of particular strain was in the contact center, and no industry was immune.
Call volume exploded during the pandemic, putting immense pressure on customer service operations. This surge brought the topic of chatbots and conversational automations to every board table in the need to bring relief to contact centers while continuing to provide high-quality customer service. The pandemic increased the adoption of chatbot technology significantly, and investment budget has been reprioritized to support chatbot initiatives, transforming chatbots from an innovative gamble to a pillar of customer service channel strategy within a 12-month period.
With significant financial investment come heightened expectations of a return on that investment. What we have observed in many companies, however, are results that fall below expectations. In particular, many companies are struggling to drive the desired volumes of queries through their chatbot solutions. And for those conversations handled by the chatbot, how sizeable is the benefit really?
After observing a range of chatbot initiatives with mixed outcomes, this article proposes a framework that successful chatbots follow in order to achieve desired levels of adoption, deliver value to end users, and ultimately drive a positive return on investment.
1. 3 Tips for your Voice and Chatbot Program from Gartner’s Customer Service Hype Cycle 2020
2. Deploying Watson Assistant Web Chat in Salesforce Lightning Console
3. Are Chatbots Vulnerable? Best Practices to Ensure Chatbots Security
4. Your Path to AI — An IBM Developer Series
Chatbots can and do provide significant end-user value when executed effectively. Some use cases might benefit from advanced NLP and contextual awareness, while others might be better addressed via button-driven dialogs. The following framework, however, proposes a model that unites all successful chatbot solutions regardless of use case, industry or channel. This model, the five P’s of Placement, Purpose, Personality, Personalization, and Product, outline challenges that successful chatbots address with clarity.
Placement
This pillar is concerned with the location and method for interacting with the chatbot. Where and how customers communicate with the solution is a key driver of volume into the funnel. It doesn’t matter how well designed your chatbot is, if no one opens the chat window and triggers the welcome message, the chatbot will fail.
Key questions to consider are:
Via what channels is my chatbot available, and specifically why have I chosen these?
- How easy is it to find the chatbot on my website (if applicable)?
- How will we drive traffic towards the chatbot?
Purpose
This pillar focuses on the reason for implementing a chatbot. While having a chatbot was just a cool thing to do before, chatbot fatigue has certainly set in amongst end users. It is now critical that chatbots are built to achieve a specific purpose. Customers are no longer satisfied talking in endless loops with a chatbot; they want to see value.
Key questions to consider are:
- What customer challenge(s) does my chatbot resolve?
- What end-user value is delivered by my chatbot?
- How does the chatbot fit into my broader channel strategy?
Personality
This pillar addresses not only the personality of the chatbot, but also how it aligns with the bot’s users. Defining an effective personality for your bot is critical for maximizing end-user engagement and pushing users down the funnel. Despite an increase in awareness of user-centric design concepts, and research highlighting the value of chatbot personality, this pillar is the one most often overlooked.
Key questions to consider are:
- What personality does my chatbot exhibit and how does this align with its specific purpose?
- How will the chatbot’s personality resonate with its end users?
- How does the chatbot’s personality reflect my company’s brand and values?
Check out The Ultimate Guide to Chatbot Personality for a comprehensive walkthrough of this topic.
Personalization
The fourth pillar assesses the extent to which conversations with the bot are tailored to the particular user. Personalization is important for increasing engagement and providing specific value to specific users. Personalization can be achieved by understanding and recalling context parameters, or presenting personalized data from back-end systems and reacting to it.
Key questions to consider are:
- How much contextual information can the bot identify and use?
- To what extent can and should the bot provide a tailored service?
- How well can the bot react to user-specific utterances such as spelling or word variations?
Product
The final pillar ensures potential outcomes of interacting with the bot are valuable ones. End users need to see value from chatbot interactions, otherwise they will not bother in the future. Value delivered may include actioning a task on their behalf, providing a recommendation, or managing an escalation to a human agent if required.
Key questions to consider are:
- What outcomes does my bot consistently deliver to the end user?
- What are the natural bundles of services my bot should provide, to limit double handling?
- How useful are the fall back options my chatbot offers?
After observing the outcomes of many chatbot projects over a number of years, the Five P’s presented above highlight the defining characteristics of a successful chatbot, regardless of use case, industry, or userbase. What should be apparent, is the critical need to place the user at the center of all development decisions. An IT-driven culture is a dominant feature of many chatbot initiatives, pursuing cool applications of the technology and losing sight of what really matters to end users. This position must be adjusted to achieve success. After all, it is the chatbot’s end users, through their adoption and repeated use of the solution, that determine the success of the bot.