The Sex Ed
Chatbot App
Come with me as we go on a sexy design journey, touching on user research, chatbot research, design processes and final prototype.
OVERVIEW
Objective: Design an app which solves a social problem
Deliverable: High fidelity prototype
Timeline: 3.5 weeks
My role: I worked independently to develop the concept, user research, personas, information architecture, job stories, user testing, branding, wireframes, mockups, prototypes (InVision), and animations (Principle).
BACKSTORY / INSPIRATION
Sex education curriculums have always been a much-debated topic across Canada. In Ontario, it was a rallying issue in the 2018 provincial election. A new, more progressive, curriculum was introduced in 2015, and it received some push back from Doug Ford, the newly elected Progressive Conservative premier, whose campaign platform included a promise to replace the new curriculum with the previous version. That said, the new improved 2015 curriculum wasn't strongly enforced. If a school board or a teacher was uncomfortable teaching the content, they had no obligation to do so. All of this tension created some glaring inconsistencies in what young Canadians were learning from school to school, leaving some students in the dark on important sexual health related information.
"Parents just don't understand!" -- DJ Jazzy Jeff & The Fresh Prince
"Parents just don't understand!" -- DJ Jazzy Jeff & The Fresh Prince
WHAT SHOULD BE DONE ABOUT IT?
Between the gaps in sexual education in schools and the misinformation found in Google search results, mixed in with actual valid resources online, I believe that students are in need of an open resource. My solution is to offer an app that allows young adults to ask sexual health related questions anonymously. I believe an app is the best solution because it puts the information directly in the palm of their hands, on the phones that they are tethered to.
With the goal of keeping things as simple and as straightforward as possible, this app could offer the following 3 functions:
-
A conversational chatbot that would answer any sex related questions a user might have, no matter how simple or complex.
-
An open forum where users can anonymously ask questions which are then answered by professionals.
-
Contact information for local support services as mentioned above, such as Planned Parenthood and Hassle Free Clinic, locations and times of various pop-up HIV/STI testing sites, information nights, etc.
USER RESEARCH
Usually for a project like this, I would interview the target audience directly; however, in this instance I felt it would be inappropriate to approach teenagers on the street and ask them about sex. Therefore, I decided to base this project on research that was already published online. Turns out there has been a lot written on this topic over the last 10 years. From sex columns in Seventeen and Teen Vogue to data by Statistics Canada to long-form research by CBS News and more, I developed a well-rounded picture of the mindset of teenagers today. Shocker: teens have a lot of questions about sex.
From this deep dive into the teenage psyche, I developed my personas & job stories:
JOB STORIES
-
SARA: When I have questions about sex, I want to be able to ask them safely and discreetly, instead of having to ask my parents.
-
EMMA: When I have questions about sex, I want to be able to find reliable information that is open and honest, so that I feel more comfortable about my body and sexual health.
-
TRISTAN: When I have concerns about my sexual health, I want to be able to reach out to someone, so that I can get support and comfort in my time of need.
It was through these job stories, that I settled on the name for the app. I was originally wrestling with the themes of a safe space or a place where the user could ask a questions as if speaking to a confidant, but through testing the names "SafeSpace" and "Confidant", they didn't garner very positive reactions. However, coming back to the job stories, I noticed the themes of openness, honesty and comfort -- thus HONESTLY was born.
TACKLING A CHATBOT - WHAT'S POSSIBLE?
When I started researching chatbots, I saw how limited most of them currently are, but also the advances that are being made, especially by Google. I quickly became aware of just how intelligent chatbots could actually be. What stood out to me was the Google Assistant Makes a Phone Call demonstration at Google I/O '18. This opened my mind to what could be possible in the near future.
I had also looked at the presentation that James Giangola from Google's Conversation Design Team gave at Google I/O '17. When building AI conversations, he focuses on the Cooperative Principle, stating, “Conversation is more than just the literal exchange of information. According to the cooperative principle, we humans naturally assume and expect there to be an undercurrent of cooperation between participants of a conversation … According to this rule of conversation, a speaker provides as much information as is necessary to advance the perceived purpose of the conversation, but not too much.”
This really was an "ah ha!" moment and it became my goal to apply the Cooperative Principle to my chatbot.
CHAT FLOW
When creating a possible chat flow for the sex-ed chatbot, I looked to the self assessment tests that Planned Parenthood Toronto’s teenhealthsource.com provides: a series of yes and no questions that offer possibilities of whether someone might be pregnant, or need to get STI testing done, etc. What I found interesting about these yes and no assessments, were the various outcomes that one could arrive at -- a quasi choose-your-own-adventure for sexual health. I mapped all the possibilities offered on the “Am I Pregnant?” assessment and ended up with the following overview:
A bit daunting at first, right?! However, I made a plan to choose one of the flows and write out a conversational chat flow, consciously being aware of the Cooperative Principle. This was thoroughly enjoyable for me since my background is in theatre/playwriting.
The result looked something like this:
Important to note is that a chatbot can only do so much. If the conversation becomes too heavy or is beyond the chatbot's capabilities, it would suggest a local support service that would be useful to the user. This was the case with this particular chat flow; the user was eventually referred to a local Planned Parenthood for her specific needs.
TONE & VOICE
Zeroing in on a tone & voice for this type of app was difficult. Sex should be fun and therefore a product that champions it should be fun as well. However, what if a user has a serious sexual health related problem that they need to address? An over-the-top and silly tone would not be appropriate. I had to find a way to strike a balance.
I decided that the chatbot could simply start off each chat interaction with “How are you feeling today?” -- the answer would guide the tone of the rest of the conversation.
This was also a way to address the separate needs of my 3 personas, who would all be coming to this app with different questions and uses.
DESIGN, MEMPHIS-STYLE
When approaching the design of the app, I wanted it to be young, friendly, accessible, and to force myself to play around with interesting colour combinations. Taking a nod from a recent article by The Cut, I zeroed in on Memphis Design and used its elements as inspiration, particularly blocky geometric shapes, block lettering next to handwritten fonts, odd angles, and hard shadows.
MOODBOARD
ITERATIONS
To allow myself to play around with various design ideas, I honed in on the app's splash page and the onboarding screens. I felt it was important to get them "right" since they would set the tone for the design of the rest of the app.
Version 1
My goal here was not to mimic Memphis Design, but to subtly allude to it.
Feedback from fellow designers was a resounding "BOO! HISS! BORING! TAKE IT FURTHER!"
Version 2
Here I wanted to see what would happen if I went full-on early '90s.
Feedback was a resounding "TOO ON THE NOSE! TOO FAR!"
In the end, I captured the essence of Memphis Design, but offset that with current styles and my own personal flair.
THE FINAL RESULT
PROTOTYPE
Putting together the prototype, I focused on the 3 main functions and implemented a similar 3-screen navigation style as Snapchat and Tinder, a structure that teenagers and young adults are used to.
If the embedded prototype is acting up (or you're on mobile), please use the following link and feel free to curse InVision under your breath: HONESTLY App on InVision
RELEASING IT INTO THE TEENAGED WILD...
Desiring some direct feedback from the target audience, I created a short Typeform survey and spread the link over social media. The responses I received were interesting -- not necessarily because they were positive or negative, but more so that it turns out teenagers do not like filling out surveys! The majority of the responses looked something like this:
A lot of "Nah", "Nope" and "Idk". LOL
Perhaps I didn't ask engaging enough questions (although I did try to keep the survey as open and straight-forward as possible). Perhaps in this particular case, multiple choice is the way to go. Although multiple choice can be seen as "leading", it would negate answers like "Nah" and "Nope".
Moral: meet with the target audience in person, engage in dialogue, ask them to explain their thoughts and, maybe most important, keep surveys away from teens!
Nonetheless, some of the results were helpful, mainly that a high majority of 16-24 year olds said they would use this app and find it helpful.
TAKEAWAYS
1. Although I gained a lot of information from my research, meeting with the target audience is necessary. Nothing can replace the one-on-one face time and the ability to probe further when things are unclear.
2. Feedback is your best friend. Be it from your target audience, fellow designers or people who fall outside of those categories -- everyone has an opinion and they might just pick up on something that you've overlooked.
3. "Accessibility first" is important. I had a "sky's the limit" mindset while working on this project, however taking accessibility into consideration in the very early stages is crucial. I've learned that going back to a "finished" product and applying accessibility best practices is a lot of work. Being cognizant of accessibility requirements from the very beginning not only saves you time in the long run, you also end up with a superior final product.
There is a lot more I could discuss, -- from animations within the app to call-to-action choices, etc -- all of which I would be happy to go through over a cup of coffee or pint of your choice. Please don't hesitate to get in touch!