Writing and Designing for Chatbots

Students will examine conversational user interfaces—that’s to say, using chat as a medium to interact with a bot.

Students will tackle key interaction-design problems around user flow, open-response vs. selected paths, and writing for voice—all through in-depth user research and testing. The student team will work with Public Good Software to research, prototype, and test an onboarding experience for the Public Good platform that uses a chat interface (or an interface that uses chat) to match users’ interests, location and passions with related nonprofits and causes. The research and planning phases of the course will explore the viability of a chat experience both as part of the Public Good homepage (or landing pages) and also as a facet of a current chat-widget embedded within media partner sites.

Learning Outcomes

Exploration of conversation UIs and best practices
Exploration of user research methods and prototyping
Exploration of successful onboarding experience

About Public Good Software

We are a marketplace for nonprofits, media companies and change-minded, engaged humans looking to make a positive impact. Our technology uses machine learning and artificial intelligence to suggest actions you can take on the causes, campaigns and organizations you care about most. Our goal is to turn important news into meaningful change.

Our founding team are startup veterans who helped launch some of the biggest online marketplaces including Sittercity, Groupon, and Orbitz. Two of our founders, Dan Ratner and Jason Kunesh, were tech and user experience leads (respectively) for President Obama’s historic 2012 reelection campaign, arguably the most successful national organizing effort ever. At Obama for America (OFA) Dan and Jason saw first hand how people organize around the causes they care about. They saw lifelong friendships form, neighbors meet neighbors, and the game changing effect of millions of small actions all pushing in the same direction.

Public Good was launched in order to put those ideas and ideals in play for people wanting to make a difference but not knowing where to start. Our goal is to transform the world of fundraising and organizing in a way that better fits the digital generations. We’re early in our journey, but we’re joined by many of the nation’s top news publishers (CNN, Upworthy, TRONC, etc.) who are now using our technology to enable their readers to take actions on the stories they read.

Faculty and Staff Leads

Joe Germuska

Chief Nerd

Joe runs Knight Lab’s technology, professional staff and student fellows. Before joining us, Joe was on the Chicago Tribune News Apps team. Also, he hosts a weekly radio show on WNUR-FM – Conference of the Birds.

Results

Design Research > Writing and Designing for Chatbots

How We Brought A Chatbot To Life Best Practice Guide

by Alex Sher | AlexLSher , Isaac Lee | IsaacRLee , Patrick Hoy , Mengyi Jenny Sun | MengyiJennySun

A chatbot creates a unique user experience with many benefits. It gives the audience an opportunity to ask questions and get to know more about your organization. It allows you to collect valuable information from the audience. It can increase interaction time on your site. In the spring of 2017, our [Knight Lab](/projects/writing-and-designing-for-chatbots/) team examined the conversational user interface of [Public Good Software](https://publicgood.com/)’s chatbot, which is a chat-widget embedded within media partner sites. Public Good’s...

Continue Reading

Project Details

2017 Spring

Important Questions

What factors (interest in causes, location, giving preferences, etc.) make the most meaningful connections between users and existing causes, organizations?
What style of chat (what tone, what language, what level of verbosity, etc.) will prove most successful at converting users or keep them engaged with the experience?
Considering the research results for 1 and 2, are there specific user types or personas that respond more or less favorably to different styles of chat.
As the platform returns results, what is the best way for users to save, favorite, etc. in order to build their profile?
Are there best practices - e.g. multiple short sessions vs. a single long one?
Do the same chat techniques map well from platform to platform - e.g. direct, Facebook, Echo, etc.?
Do users want a more transactional approach or a more personality rich experience?

Sample Milestones

Weeks 1-2 Ramping up Understand PGS goals, platform, onboarding goals
Weeks 3-4 Develop project plan and testing approach
Weeks 5-6 User Interviews and other research methods
Weeks 7-8 Prototype development, testing, iteration
Weeks 9-10 Final prototype delivery and presentation

Outcome

A high fidelity prototype powered by ‘working’ code or simply a functional design prototype accompanied by a document outlining research, user feedback, and testing results.

Learning Outcomes

About Public Good Software

Faculty and Staff Leads

Joe Germuska

Chief Nerd

Results

How We Brought A Chatbot To Life Best Practice Guide

Project Details

2017 Spring

Important Questions

Sample Milestones

Outcome

Students

Patrick Hoy

Isaac Lee

Alex Sher

Mengyi Sun