What are rule based chatbots
Rule based chatbots guide client requests with fixed options based on what they are likely to ask, they then provide fixed responses. Rules based chatbots are limited to basic scenarios that sometimes lead to frustrating experiences.
What are AI, machine learning chatbots
AI Machine Learning chatbots, the new generation chatbots can engage in natural conversation, for example speak with your brand tone of voice or use local dialect terms – you may hear this referred to as natural language processing.
What type of chatbot should my law firm use, rule based or AI, machine learning chatbot?
Let’s look at two types of chatbot, rule based or AI, machine learning. Let me take you through some pros and cons of both systems.
The diagrams below illustrates the two systems, left to right, the Rule Based Chatbot and AI, Machine Learning Chatbot.
Rule based chatbots “the pros”
Firstly creating a rule based chatbot is quicker and simpler than an AI, Machine Learning chatbot. This is because a rule based chatbots give answers to your client’s questions from a set of predefined rules you create from known scenarios. For example a chatbot will present your firms service options, the client then select which they want. A simple decision tree is then used based on conditional statements. This makes them quicker to deploy, gets them up to speed quicker and has a lower cost of development.
Unlike AI, machine learning chatbots, rule based chatbots are simpler and don’t need to learn overtime to understand your client’s needs to provide the correct answers for them.
Rule based chatbots do have some advantages over AI, machine learning chatbots but they also have short comings that need to be fully considered.
Rule based chatbots “the cons”
Rule based chatbots can’t learn on their own, they only provide answers your legal team provides from a predefined set of rules. That means your clients experience with rule based bots is very linear. In other words if your client asked questions outside its preset understanding they fail and need human intervention.
Rule based chatbots can’t offer a personised experience, for example if you gave a chatbot your name it won’t be able to remember it. As people inevitably use different grammatical structures, rule based chats breakdown. For example when a client wants a Will, they may say, “I want to start a new Will,” another client may use a different grammatical structure, “I need Will for myself.” We as humans understand both requests but a rule based chatbot can’t work that out and then has to pass the chat to a lawyer that in turn frustrates your client.
AI, Machine Learning chatbots “the pros”
Let’s now look at the pros of AI, Machine Learning chatbots – their biggest advantage over others is they are self learning and can be programmed to communicate in your brand voice and even local dialect.
That said it would be wise to consider the time it takes to train them. Think them as real people starting from scratch and you get the idea. In return you gain a legal expert who works 24 hours a day and can do all the mundane tasks where we humans are too expensive. If you have lots of data for them to work with they can learn from it and that will save your law firm time and money.
AI, Machine Learning chatbots engage in end to end client requests and provide services without human interaction with multiple consecutive conversations 24 hours a day.
They can also be developed to understand different languages, dialects and can personalise communications with your clients where rule based chatbots can’t. They understand intent, emotions and can be empathetic to your client’s needs.
As opposed to rule based chatbots AI, Machine Learning chatbots provide a more human experience, they can:
- Chat in natural conversation
- Workout misspellings
- Keep improving over time
- Available to clients 24 hours a day 7 days a week
AI, Machine Learning chatbots “the cons”
For sure AI, Machine Learning chatbots are very cleaver, but their shortcomings are around context when communicating with us humans. By that I mean, we automatically change how we talk with young people v more formal tones with clients. Given chatbots can’t understand that context they communicate the same way regardless of what age or gender of the person.
AI, Machine Learning chatbots are created using Natural Language Processing which is in great demand in customer facing applications. It’s worth noting this does need time programming and training if law firms create them from scratch.
Once created law firms then need to keep it updated with any changes or queries that’s may have been missed. It’s always good to keep testing and reviewing to make sure it’s does what you were expecting to do.
If you are interested in learning more about Artificial Intelligence and Machine Learning chatbots we’d love to discuss how they can help your law firm.