The 3 Types of Chatbots & How to Determine the Right One for Your Needs
Chatbots come in three distinct flavors. Make sure to choose the one that’s right for you.
Don’t bring a bazooka to a water gun fight 🔫
If you think that all chatbots are created equally, you’re unfortunately in the wrong. Chatbots today come in all shapes and sizes and have varying levels of capabilities. While basic chatbots may be adequate for most scenarios, some scenarios require more advanced chatbots.
On the flip side, don’t go overboard and build an elaborate chatbot with AI capabilities that rival those of IBM’s Watson if your chatbot is merely helping users to determine what wine to pair with their dinner. In that instance, a chatbot based off of clickable menu buttons or simple keyword recognition would suffice.
To learn more about balancing AI capabilities with UX design when building chatbots check out one of my previous posts on the subject.
Menu/Button-Based Chatbots 📍
Menu/button-based chatbots are the most basic type of chatbot on the market today. In most cases, these chatbots are glorified decision tree hierarchies presented to the user in the form of buttons. Similar to the automated phone menus we all interact with on almost a daily basis, these chatbots require the user to make several selections to dig deeper towards the ultimate answer.
While these chatbots are sufficient for answering that handful of nagging FAQs that make up 80% of support queries; they fall well short in more advanced scenarios in which there are too many variables or too much knowledge at play to predict how users should get to specific answers with confidence. It’s also worth noting that menu/button-based chatbots are the slowest in terms of getting the user to their desired value.
Keyword Recognition-Based Chatbots ⌨️
Unlike menu-based chatbots, keyword recognition-based chatbots can listen to what users type and respond appropriately, or at least try to. These chatbots utilize customizable keywords and AI to determine how to serve an appropriate response to the user.
For example, if a user asked the question ‘How do I set up an auto logout transaction on a Poynt device?’, the bot would likely use the keywords ‘auto’, ‘logout’, and ‘Poynt’, to best determine which answer to respond with.
These types of chatbots fall short when they have to answer a lot of similar questions. The chatbots will start to slip when there are keyword redundancies between several related questions.
It is becoming quite popular to see chatbots that are a hybrid of keyword recognition-based and menu/button-based. These chatbots provide users with the choice to try to ask their question directly or use the chatbot’s menu buttons if the keyword recognition functionality is yielding poor results or the user requires some guidance to find their answer.
Contextual Chatbots 🧠
Contextual chatbots are by far the most advanced of the three bots discussed in this post. These chatbots utilize Machine Learning (ML) and Artificial Intelligence (AI) to remember conversations with specific users to learn and grow over time. Unlike keyword recognition-based chatbots, contextual chatbots are smart enough to self-improve based on what users are asking for and how they are asking it.
For example, a contextual chatbot that allows users to order pizza will store the data from each conversation and learn what the user likes to order. The result is that eventually when a user chats with this chatbot, it will remember their most common order, their delivery address, and their payment information and merely ask if they’d like to repeat this order. Instead of having to respond to several questions the user just has to answer with ‘Yes’ and pizza is on its way!
For a contextual chatbot to be useful, a data-centric focus is imperative. To learn more about this data-centric approach to building AI and chatbots check out one of my previous posts on the topic.
While this pizza ordering example is elementary, it is easy to see just how powerful conversation context can be when harnessed with AI and ML. The ultimate goal of any chatbot should be to provide an improved user experience over the alternative of the status quo.
This improvement in user experience often arises from providing particular value such as the delivery of a hot, delicious pizza in less time than before. Leveraging conversation context is one of the best ways to shorten a process like this via a chatbot.
So, which one is right for you❓
When deciding which chatbot is right for you place yourself in the shoes of your users and think about the value they’re trying to receive. Is conversation context going to significantly impact this value? If not, then it is probably not worth the time and resources to implement at the moment.
Another thing to consider is your target user base and their UX preferences. Some users may prefer to have the chatbot guide them with visual menu buttons rather than an open-ended experience where they’re required to ask the chatbot questions directly. All the more reason to have users extensively test your chatbot before you fully commit and push it live.
The right chatbot is the one that best fits the value proposition you’re trying to convey to your users. In some cases, that could require enterprise-level AI capabilities; however, in other instances, simple menu buttons may be the perfect solution.
Do your research and ultimately make the best choice for your users. After all, unsatisfied users rarely ever equate to repeat users…
