A chatbot is a form of artificial intelligence (AI) that talks with people, provides information, and performs tasks. Contrary to a fad like bellbottoms, AI is becoming more entrenched in our lives as it becomes more useful. For example, AI is already embedded in the gadgets and software we use everyday, like Apple’s Siri, Microsoft’s Cortana, and Amazon’s Alexa.
As an iPhone user, I know from experience that Siri can be useful, sometimes. Other times, She’s pretty useless. Her biggest problem might be that she tries to do too much; therefore, she does a mediocre job at many tasks. Yet, she’s getting better. As Siri improves, I’m sure I’ll use her more and more.
Similar to Siri, a chatbot is an AI assistant for website users. However, unlike Siri, you can control how a chatbot is being used on your website to make sure it’s helpful. Rather than doing a mediocre job at many tasks, a better strategy is to limit a chatbot to a few specific tasks it can handle very well. For instance, a chatbot would not be a good replacement for a human live chat agent (algorithms simply aren’t as good at holding conversations as humans, yet), but a chatbot might be even better than a human in other cases.
Below are three examples of how chatbots can effectively assist website users.
3 Use Cases for Chatbots on Credit Union Websites
Guide Users Through Processes
I know several credit union marketers who really want to put chatbots on their websites after meeting Chip the chatbot (view the demo to see what I’m talking about). These excited marketers imagine Chip guiding a person through the process of getting an auto loan, becoming a member, or opening a new account. A conversation with Chip might go something like this:
The chatbot stores this conversation in its database. Next time the user returns to the website, the chatbot remembers the user (using cookies) and the conversation can continue where they left off. That’s one thing a chatbot can do better than a human live chat agent: a chatbot can remember every user and every conversation. Plus, the chatbot is on the website 24/7/365; it won’t even complain about working on Christmas 😉
Recommend Products and Services
Another use case for a chatbot is recommending products and services. This use case really clicked for me while talking with a credit union marketer named Kari. Kari says her credit union is using a new approach where they get members to complete a survey that asks about financial needs. Once the survey is complete, an employee can recommend products and services based on the member’s responses. This approach is really helpful for members and effective for the credit union. A chatbot is the perfect way to implement this same approach on a website:
A chatbot can also be used to answer Frequently Asked Questions. Recently, my friend Nancy and I were talking about questions that members frequently ask. She says people constantly ask for their routing number, so Nancy wants to improve her website’s search functionality to make finding the routing number easy. Making the routing number show up first in the website’s search results is a fine solution, but getting that information from a chatbot might be even more simple and delightful for users:
As with any technology, you have two options: build or buy.
Build Your Own Chatbot
If you google “chatbot for website” you’ll find all sorts of software services for building chatbots. Below are a few things to keep in mind as you evaluate these many software platforms.
All Chatbots Are Not Created Equal
Today’s chatbots fall on a spectrum. On the left of the spectrum are simple chatbots designed to facilitate human-to-human communication. Simple chatbots answer questions, capture information and introduce a human only when necessary. On the right of the spectrum are Artificial Intelligence (AI) chatbots designed to mimic human-to-human communication.
How to Build a Simple Chatbot
If you know how to use the internet you can build a simple chatbot for your credit union in minutes. The easiest way to create a simple chatbot is to use a platform with a visual builder. The best tool we’ve found for building simple website chatbots is Motion.ai.
How to Build an Artificial Intelligence (AI) Chatbot
If you can code, you can wield the amazing powers of artificial intelligence and natural language processing. To build Chip the chatbot we use the AIaaS platform. AIaaS is an artificial intelligence API which processes natural language (e.g., “What is my routing number?”) and returns the most intelligent response available. A developer will have no problem using a platform like this. Other amazing platforms for building artificial intelligence chatbots are IBM Watson, Microsoft Bot Framework, api.ai and wit.ai. If you aren’t a software developer, you’ll need to work with a developer to use these advanced platforms and deploy a chatbot on your credit union website.
Chatbot technology is readily available. What’s needed are creative people who can figure out how to use chatbots in meaningful ways for credit unions. While the services mentioned above give you the software to build a chatbot, you have to create the content that fuels conversations. Creating and organizing the content is, perhaps, the greatest challenge you’ll face in building your own chatbot.
If you’ve built or are building your own chatbot, tell us about it in a comment below!
Buy a Chatbot Solution
Though chatbot technology is readily available, there aren’t many chatbots that have been taught to chat about topics relevant to credit unions.
In fact, I know of only one chatbot that’s built specifically for credit union websites: Chip the chatbot. (Disclaimer: Chip is built by my company, BloomCU. Nevertheless, I really don’t know of any other chatbots built for credit union websites.) There are several ways to use Chip to engage users of credit union websites.
Personetics also makes a chatbot for credit unions, but rather than it being used on websites, it works on social platforms like Facebook Messenger and Skype. It’s called Personetics AnywhereTM and appears to mostly perform money and account management functions. For example, a member can message the chatbot, “How much did I spend last month?” and the chatbot would respond, “You spent $2,000.” It seems similar to text banking.
There aren’t many chatbot solutions for credit unions yet because most people are unfamiliar with chatbots and how to use them. I believe we’ll see more solutions for credit unions in the years to come. For now, chatbots are for those akin to Dr. Frankenstein who want to give their websites life and marketing futurists who want to embrace the future of website technology now, in 2016.
More Ways to Use Chatbots?
In a sense, chatbots bring websites to life. Instead of a website that sits idly while you browse, a chatbot can give a website personality. Chatbots can . . .
- Talk with you
- Remember your name
- Recall the webpages you visit
- Initiate conversations based on your interests
- Help you find information
- Answer your questions
- Give you guidance about your financial goals
There are probably endless ways to use chatbots for credit unions. I’ve shared a few use cases and you can find a few more ideas here: https://bloomcu.com/chatbot/. How would you use a chatbot on your credit union website? Share your ideas in the comments below.