Four tips for introducing an FAQ chatbot
- Laura Dernie
- Archive
- August 5, 2020
- Updated:
Table of Contents
Getting your own chatbot up and running is not as tricky as you may think. In fact there are many no-code options available and you can be setup in minutes. We have put together a list of tips to consider before you jump in on augmenting your customer service.
1. Talk to your teams
Involving any customer-facing member of your team in the creation of your FAQ chatbot ensures that team members know they are not being replaced, as well as helping you get the best content for the chatbot. They know better than anyone the questions they get all too often, so they will see the value in not having to answer these question and being able to focus on other tasks.
If you don't know where to start, try searching your company name to see what people are asking on forums. You can always add relevant question and answers later as people's needs become more clear.
2. Give your chatbot a personality
In any business, it is important to understand your customer personas and the journey that they go through as a buyer. You can strengthen your customer personas by talking to different members of your team to get the full picture and develop your user journey. You can now use this perspective to fine-tune your FAQs and create a conversational experience that your customers will enjoy.
3. Define your end game!
No, we are not talking Avengers, we are talking about what you want to achieve. Getting people's attention is hard, so when they come to you with a question, don't miss the chance to push for your end goal! Make sure to include suggestions or calls to action at the end of a conversational flow. You may be able to encourage a sale directly from the chatbot.
4. Augment customer service
While chatbots have many perks, like instant customer service and improving customer engagement, they don't always have all the answers, and some people just want to speak to a human. Letting your customers know that a human option is available releases this tension. Customers are more likely to give the chatbot a go, knowing that they can speak to a real person if it doesn't get them the answer they need.
FAQs
Why chatbot introductions fail — and how to avoid it?
FAQ chatbots have a mixed reputation in business circles, largely because many early deployments were done badly: limited knowledge bases that could not answer common questions, bot personalities that frustrated rather than helped, and no clear pathway to escalation when the bot could not help. Businesses that had bad experiences with first-generation chatbots are understandably sceptical.
The good news is that the technology has improved significantly, and the lessons from failed deployments are well understood. The four tips in this guide are a distillation of what distinguishes successful chatbot deployments from unsuccessful ones — and they are all about implementation and management, not about the technology itself.
Is the knowledge base everything?
An FAQ chatbot is only as good as the knowledge it has access to. A knowledge base that is accurate, comprehensive, and regularly updated will produce a chatbot that genuinely helps users. A knowledge base that is stale, incomplete, or poorly organised will produce a chatbot that frustrates users and damages trust.
The most important investment in a successful chatbot deployment is not the technology — it is the time spent building and maintaining the knowledge base. This means identifying the questions that customers and users actually ask (not the questions you think they ask), writing answers that are clear and accurate, and establishing a regular review process to keep the knowledge base current as products, processes, and policies change.
Measure what matters?
The metrics that matter for an FAQ chatbot are the ones that reflect user outcomes: containment rate (the proportion of queries that are successfully resolved by the bot without escalation), customer satisfaction scores, and the reduction in human agent workload. These are business metrics, not technology metrics.
Monitoring these metrics and using them to drive improvements to the knowledge base and bot configuration is what transforms a chatbot from a one-time deployment into a continuously improving customer service capability.
