A gentle introduction into chatbot creation
For healthcare — a simple pain survey bot
What are chatbots?
Chatbots are interactive applications designed to simulate human conversation. They can be found in a number of different sectors from retail to banking to online shopping and more.
The chatbot market is expected to reach $1.25 billion by 2025 with a compound annual growth rate (CAGR) of 24.3%. That figure is only expected to grow as more companies begin to acknowledge their potential.
Why build a chatbot for healthcare?
Chatbots offer numerous advantages for the healthcare industry. Applications can range from providing medical reminders or feedback, scheduling appointments, and improving access to medical information. Beyond that, they can also be useful, objective tools for clinical data collection and research surveys.
What we will create
Check out the finished product here.
Start
Okay, let’s start building our own chatbot. We will use landbot.io for this tutorial. Landbot offers an easy-to-use interface for building rule based chatbots for free (up to 100 chats per month). This means no coding is needed and no cost included, which is great for people dipping their toes in the chatbot water. We will use landbot’s logic-building user interface to make a pain survey.
Follow these steps to begin:
- Create a landbot.io account
- Select -> create a new bot
- Select -> web-based bot
- Select -> create a bot from scratch
Build
- Enter introductory texts and an open ended question.
2. Connect the string of texts to a yes/no question to assess if the participant is experiencing any pain.
3. Customize icons and text.
4. Create responses based on participant selection.
5. Add a question asking the participant to describe his or her pain.
6. Customize the question and create a variable to collect the data.
7. Create a scale to assess the intensity of pain.
8. Add a question about pain relief methods.
9. Loop to the end of the survey.
10. Add human takeover function and assign an agent.
11. End the survey by closing out of the chat.
Design
Landbot let’s you control the look and feel of the chatbot by customizing the color scheme, layout, icons, etc.
Share
Use the HTML code to create a landing page.
Analyze
Analyze the responses by viewing user submitted data.
Conclusion
That’s it! We created a conversational agent to assess pain. If you would like to see the pain bot in action, click here. Hopefully this article provides a brief glimpse into the potential of chatbots and the future of digital health. Other applications could include mental health aids, therapy suggestion bots, medical provider referral agents, and beyond. Go on and create something great!
