Ava

Year

2022

Role

Conversation Designer
Product Designer
Chatbot Developer

Industry

Health Tech
AI

Conversational Agent for a Mental Health App

— Overview

With the advancement of technology and the existence of AI, understanding audiences with the use of chatbots has become easy and very lucrative for a lot of businesses.

These digital assistants elegantly hold interactions between people and services by strengthening customer experiences. Apart from the cost savings companies will attain from chatbots, they have also been efficient in helping customers get answers to their questions in real-time.

Businesses are looking to make information more accessible and effective by incorporating chatbots, which help achieve quality support and solutions. Industries, especially the health sector, have remodeled to become more automated.

— The Value

Creating an engaging experience for people coping with mental health issues

Conversational AI for the healthcare industry has made access to health care or medical support during emergencies easy. The introduction of these chatbots enhances the efficiency as well as trust of both doctors and patients.

Mental health chatbots offer relevant tools and resources that help users cope with their health, which can be accessed from the comfort and safety of their homes. Because of the anonymity of these chats, people will feel more comfortable sharing their worries and deep fears.

So, not only are mental health chatbots making waves in the healthcare industry, but chatbots are also emerging to serve new and beneficial purposes in the world.

— About Ava

Your trustworthy conversation partner

Ava is a chatbot designed for people who are coping with mental health issues. The chatbot is capable of forming bonds with users by talking about their troubles and receiving counsel from the bot.

With Ava, users can set a time of the day when the bot can check in on them about their wellbeing or send them funny videos.

Sometimes you just need to talk to someone who will listen to you without passing judgment and help you make sense of things. This is where Ava enters the picture. 

While designing the chatbot, I made sure to keep some limitations in mind and what it is designed to do.

— Bot Identity

Name of Bot : Ava
Gender : Female
Age : 30
Character Traits: Delightful, witty, emphatetic, optimistic
Interaction Goals: Trustworthy, support, seamless
Tasks
  • Maintain conversations with users around mental wellbeing.
  • Give basic advice on how to improve their mental well-being.
  • Helps users understand and manage their mental state.
  • Connect with mental health professionals when necessary.
Chatbot Limitations:
  • Ava the chatbot is a conversational partner, guide, or coach. It is not replacing a human therapist, so it cannot imitate human conversation.
  • It cannot be used as a diagnostic tool.

— Design Process

A. User Limitations and Needs

Understanding the users and their goals is the first and most important thing I do when designing a product. Due to the stigma surrounding mental health issues and the cost of affording therapists, people are hesitant about getting proper mental health care.

People dealing with mental health issues need a place to find relief and comfort from their anxiety, burnout, and depression without feeling like a bother, which is why conversational chatbots are important because they are always ready to talk to you. That being said, these chatbots are readily accessible for people seeking psychological relief and setting the right personality and tone can go a long way towards achieving their purpose, which is to make a great companion.

B. User Goals

  • To gain support and guidance with coping with mental breakdown and improvement.
  • To gain better understanding of their mental states on their own.
  • To gain a friendly ear to listen and provide instant and available chat.

C. Other Likely Limitation and Solution

Because of the chatbot's limitations, it may be unable to recognize when a user is in crisis and provide an instant connection with a human therapist. As a result, I've added an SOS button to the top of the chat screen. This allows users to speak with a therapist immediately when they need to, with or without having to deal with the bot.

D. Designing for Security and Data Protection

When working on products with sensitive data, it is important to make sure that this data is protected. This action enhances trust between users and products/businesses.

An effective solution for this is end-to-end encryption of the user and chatbot interactions or an anonymous user profile on the database.

— Sample Dialogue & Persona

User Persona: Nancy is a 24 year old masters degree student passionate about becoming a business analyst.
User Context: Nancy is grieving the death of her sister
User Goals: she wants to find relief by speaking to someone about how she feels
Key Use Case: find relief, emotional support

Scenerio A

Scenerio B

— Chatbot Scripts

Due to the chatbot's conversational nature, not every conversation was included to chat scripts because it would be challenging to do so.

Click here to get a clearer view of chatbot scripts.

— Voiceflow Prototype/Demo & User Testing

The chatbot's short demo/prototype and user testing are shown below.

Voiceflow Prototype

After I designed the prototype and shared a link to it for testing, I received evaluative feedback on what to improve. I worked on this feedback which I marked as evaluated and iterated, and it all helped me make the chatbot feel more human rather than robotic.

— What Factors Will Impact Product Success?

  1. Intent-based metrics
  2. Great UI/UX
  3. User satisfaction and returning rates
  4. Continuous AI training

— Conclusion

I had a good time working on this project. During my research, reading, and working on this project, I learned the importance of intents and entities in chatbot design, as well as how positive emotions and empathy can improve the quality of the user experience and business goals of products in all sectors and industries.

Importance of intents and entities in designing chatbots
  1. Without appropriate intent programming, a chatbot will not evolve.
  2. Intents and utterances in chatbots are both important in providing what customers want and need.
  3. With the use of natural language processing (NLP), users can have a more engaging and smooth conversation with a chatbot in a similar way that they would with a human.

I'm fortunate to have stumbled into the fascinating world of conversation design and by the end of this project I learned how to use Voiceflow.

Next Project
Juice Up