designing conversations: Conversational interfaces, Bot Interactions, Chatbot as personalities (UX 전략적 통찰과 방향성 중심으로)

Designing
Conversations:
Conversational interfaces,
Bot Interactions,
Chatbot as personalities
(UX 전략적 통찰과 방향성 중심으로)
2016
Billy(최병호)/BillyChoi@Gmail.com
중앙대학교 교수
홍익대학교 영상대학원(HCI개론 강의)/
연세대학교 공학대학원(서비스디자인경영 강의)/
성균관대학교 일반대학원 휴먼ICT융합학과(교수)/
HEDcentric UX미래융합전략연구소(연구소장)
InnoUX(대표이사)
Research Data: http://www.slideshare.net/BillyChoi/
Blog: http://blog.naver.com/soularchitec
Twitter/Facebook: ILOVEHCI

© 2016 Billy All rights reserved.Designing Conversations: Conversational interfaces, Bot Interactions,
Chatbot as personalities(UX 전략적 통찰과 방향성 중심으로)
Table of Contents
• Basecamp Innovation and Insights
1. Chatbot and Brand experiences
2. Definition and Strategy
• Delusions and Insights
1. The more the better for conversation?! The wisdom from the social science!!
2. Computers? Social Actors?! Welcome to Similarity-attraction world.
• Deliberation and Insights
1

Basecamp Innovation 1.
Chatbot and
Brand
experiences

Here a bot, there a bot, everywhere a bot
Reference: http://www.message.io/bottranslation

Here a bot,
there a bot,
everywhere a bot
References:
• http://www.message.io/bottranslation
• http://www.slideshare.net/PatrickCullen2/chatbot-interfaces?qid=99d3c408-6f18-40a2-a373-4c5504939949&v=&b=&from_search=10

References:
• https://willowtreeapps.com/blog/part-2-shortcuts-to-chatbot-emotional-intelligence/
Make sure to fit the tone to the task at hand,
as well as the brand your bot is representing.

References:
• https://www.punchkick.com/blog/2016/10/05/how-bots-fit-into-your-apps-ux
Make sure to fit the tone to the task at hand,
as well as the brand your bot is representing.
Brands need to identify a place where a
conversational UI can provide real value to users,
and that may be within the confines of an app or
website, and it may be outside of them.

Basecamp Innovation 2.
Definition and
Strategy

Reference: http://www.slideshare.net/EnjoyDigitAll/are-chatbots-customer-relation-future-infographic-made-by-enjoydigitall?qid=99d3c408-6f18-40a2-a373-
4c5504939949&v=&b=&from_search=24

Reference: http://www.slideshare.net/Warply/chatbot-workshop-how-to-build-onedigitized16

Reference: http://www.slideshare.net/TeeWeeAng/ai-assistant-and-chatbot-trends-for-enterprises?qid=99d3c408-6f18-40a2-a373-
4c5504939949&v=&b=&from_search=1

Reference: https://twitter.com/MikeQuindazzi/status/786189180573483008

Reference: https://uxdesign.cc/impact-of-chatbots-personal-assistants-on-uxd-2072ab749b5#.qvfuhicfx
Context
Scenarios
Culture
BM with UX Model
Social Trust
Contextual Design
Scenario Design
UX Modeling
Change Habits
Design the Trust Model
Natural UI
Zero effort (and
connected) UI
Craft a Personality

References:
• http://www.slideshare.net/Warply/chatbot-workshop-how-to-build-onedigitized16
• https://uxdesign.cc/impact-of-chatbots-personal-assistants-on-uxd-2072ab749b5#.qvfuhicfx
Our job as an Experience
Designer would be to predict
the rate of change then
anticipate the behaviour of
user and how it might change
with the introduction of new
revolutions (machines, voice
interaction systems,
conversational bots, BCI etc.)
And, therefore, create
groundbreaking products.

“Use” Users For UX
“Buy” Customers For CX or ServiceDesign For Customer eXperience or Service experience
Design For User eXperience or (Interactive) product experience

“Buy” Customers For CX or Service
“Talk” Player For PX(Playful eXperience)(Saying/Typing and Respond)
Design For Customer eXperience or Service experience
Design For Player’s Playful eXperience
(embodied (interactive) product experience and service experience)

“Buy” Customers For CX or Service
“Talk” Player For PX(Playful eXperience)(Saying/Typing and Respond)
“He has a
magic lamp
with a genie
inside, who
grants wishes.”
Design For Customer eXperience or Service experience

“Talk” Player
(Saying/Typing and Respond)
“He has a
magic lamp
with a genie
inside, who
grants wishes.”
For PX(Playful eXperience)
• Creating a magic lamp with a genie(Craft a Personality);
Creating a genie-like UX
• Personalized god in a box; Era of IPA(Intelligent Personal
assistant)
• Conversation-as-a-Platform(CaaP); Software-as-a-Service
(SaaS); Platform-for-Everything
Designing Conversations:
Conversational interfaces, Bot Interactions, Chatbot as
personalities

It’s not all about the chat; Define Chatbot Domain
19
Reference: https://chatbotsmagazine.com/how-i-built-and-launched-my-first-chatbot-in-hours-764d070242e5#.ater75l3o
How broad is the understanding?
Can the chatbot find any type of music
studio and also any location in the
world? Or just music studios in the
United States? Or only music studios in
Los Angeles?
How deep is the knowledge?
Will the chatbot know of every music
studio that’s on Studiotime? Or just the
Top Line music studios? Will it share
high quality photos of the studio if
someone wants to see more photos?

Reference: http://www.slideshare.net/InteliWISE/inteliwise-chatbot-for-facebook-10-2016?qid=99d3c408-6f18-40a2-a373-
4c5504939949&v=&b=&from_search=2

Reference: http://www.slideshare.net/mmatty5/conversation-design-from-zero-to-hero-memories-of-a-bot-trainer?qid=99d3c408-6f18-40a2-a373-
4c5504939949&v=&b=&from_search=11

References:
• http://www.slideshare.net/mmatty5/conversation-design-from-zero-to-hero-memories-of-a-bot-trainer?qid=99d3c408-6f18-40a2-a373-
4c5504939949&v=&b=&from_search=11
• https://www.facebook.com/whatsInMyTech2.0/photos/pcb.334546823564509/334546463564545/?type=3
?
암환자가 스스로 자신의 다양한
데이터를 지속적으로 입력하여
생명 보존이 가능한 UX이란?
!
Chatbot & Toy
Kirobo Mini is a new companion robot from Toyota

Delusions and Insights 1.
The more the
better for
conversation?!
The wisdom
from the social
science!!

Avoid nagging users with unhelpful or irrelevant messages
• Version 1.0*
 Every time a user typed “Dear . . . ,” Clippy would dutifully
propose, “I see you are writing a letter. Would you like some
help?”—no matter how many times the user had rejected this
offer in the past.
 Clippy would give unhelpful answers to questions, and when
the user rephrased the question, Clippy would give the same
unhelpful answers again.Image source: http://bit.ly/1PDos4G
• Source: Clifford Nass & Corina Yen, 2010
• Reference:: https://www.punchkick.com/blog/2016/10/05/how-bots-fit-into-your-apps-ux

Scapegoating(Create a scapegoat; 죄를 뒤집어씌우는 전략) (1/2)
• Strategy*
 Social science literature to find simple tactics that unpopular
people use to make friends.
 Without any fundamental change in the software, the right
social strategy rescued Clippy from the list of Most Hated
Software of All Time; creating a scapegoat bonded Clippy
and the user against a common enemy.
Image source: http://bit.ly/1PDos4G
* Source: Clifford Nass & Corina Yen, 2010

Scapegoating(Create a scapegoat; 죄를 뒤집어씌우는 전략) (2/2)
• Version 2.0*
 After Clippy made a suggestion or answered a question, he
would ask, “Was that helpful?” and then present buttons for
“yes” and “no.” If the user clicked “no,” Clippy would say,
“That gets me really angry! Let’s tell Microsoft how bad their
help system is.”
 He would then pop up an e-mail to be sent to “Manager,
Microsoft Support,” with the subject, “Your help system needs
work!” After giving the user a couple of minutes to type a
complaint, Clippy would say, “C’mon! You can be tougher than
that. Let ’em have it!”
Image source: http://bit.ly/1PDos4G

착각과 통찰
• The More the Better(대화 요청 빈도와 시기, 대화의 양, 대화의 속도 등) & Less is
More(사용자의 컨텍스트 추적 기반의 알고리즘적 접근, 일정 수준의 정해진 행동패턴을
반영한 자동 처리) for conversation?!
• The wisdom from the social science!!; 사회심리학적 지식의 응용 필요
28

Watch your message pacing
• 아내의 속도에 맞게 대화를!
Our A.I. can type a thousand words per minute, but that’s not what people want from a
chat interface.
• 아내의 호흡(사용자의 특성)에 맞게 탄력적으로(학습을 통해) 대화를!
We intentionally pace how fast a user receives our messages to make the experience feel
more natural; Pace messages at human reading speed. If your bot blurts out too much text
instantaneously, this can be jarring for users to keep up with.
We’ve tested a lot of different levels of speed and found that adding a .02 second delay
helps with engagement.
We hope to create a feature that will analyze the way a user interacts with our system and
adjust the pacing for each individual, and we already have enough data to appropriately
adjust the pace by age.
29
References:
• http://venturebeat.com/2016/07/27/5-tips-for-writing-chatbot-scripts/
• https://chatbotsmagazine.com/top-6-conversational-skills-to-teach-your-chatbots-ec4eb019a23d#.xt1h59xhu

Limit the length and number of messages
• Engagement drops with every line of text over three lines, which we call the
“glanceable tipping point.”
They need to be invested in the answer they are about to receive. If they ask for
advice from a local doctor who accepts their insurance, they will take the time to
read a long message because the information matters to them. We’ve also
noticed that users don’t like receiving too many messages in a row without a
break.
• We’ve now added a user input option after 4-5 messages to break up the text and
give the user a few seconds to catch their breath. Even a simple response like “OK” or
“cool” works to pace the influx of texts.
• Keep messages short, sweet, and readable!
30
Reference: http://venturebeat.com/2016/07/27/5-tips-for-writing-chatbot-scripts/

Delusions and Insights 2.
Computers?
Social Actors?!
Welcome to
Similarity-
attraction world.

References:
• http://www.slideshare.net/Warply/chatbot-workshop-how-to-build-onedigitized16
Add personality to your bot using appropriate
avatars, names, biographical background, and other
humanizing assets

Computers Are Social Actors
• Humans expect computers to act as though they were people and get annoyed
when technology fails to respond in socially appropriate ways.(컴퓨터가 사람처럼
행동하길 기대하고, 기술이 인간적인 방식으로 반응하지 않는 것에 불만을 가짐)
Show that people treat computers as if they were real people(컴퓨터와 미디어를
사람처럼 대하는 사람들의 성향 관련 연구)
 Nass, C., and Brave, S. B. (2005). Wired for speech: How voice activates and
enhances the human computer relationship. Cambridge, MA: MIT Press.
 Reeves, B., and Nass, C. (1996). The media equation: How people treat
computers, television, and new media like real people and places. New York:
Cambridge University Press.

Similarity-attraction(유사성-매력 전략)(1/4)
• Personality traits(성격 특성); 5살 무렵에 형성
• Four Personalities
34
비판형 외향형
내향형 수용형
냉정형 다정형
순응형
지배형
협력

• This is a reproduction of one of the most famous of the Tiffany stained-glass pieces—
the colors are absolutely sensational! This first-class, handmade copper-foiled stained-
glass shade is over six and one-half inches in diameter and over five inches tall. I am
sure that this gorgeous lamp will accent any environment and bring a classic touch of
the past to a stylish present. It is guaranteed to be in excellent condition! I very highly
recommend it.
• This is a reproduction of a Tiffany stained-glass piece. The colors are quite rich. The
handmade copper-foiled stained-glass shade is about six and one-half inches in
diameter and five inches tall.
35

• You should definitely select option A instead of option B. There are at least six
reasons why this is the right option. I am 90 percent confident of this assessment.
• Perhaps you should select option A instead of option B? It seems like there are
reasons why this might be the right choice. I am 40 percent confident of this
assessment.
36

• Similarity-attraction affects people to such a degree that they feel positive toward
not only similar people but also anything associated with those similar people. For
example, in the experiment, not only did participants like the sellers who were similar
to themselves, they also felt more positive about the items associated with the
similar sellers.(유사성-매력 효과는 긍정적 감정 뿐만 아니라 유대감 유발. 심지어
성격이 비슷한 판매자가 경매에 올린 제품까지 선호)
• 외향성 음성과 내향성 음성 동일 적용; 음량, 음역, 음성 속도; 성격과 음성의 일관성
중요하게 판단함
• When the introduction to a computer-based “Entertainment Guide” matched users’
personalities, users found the recommended music to be significantly better, even
though the recommendations themselves were identical.(동일 음악을 추천하여도
서비스 도입부가 자신의 성격에 부합하면 선호 발생)
37

Reference: Dormehl, Luke (2014-04-03). The Formula: How Algorithms Solve all our Problems … and Create More. Ebury Publishing.
Quantified Self movement
Self-knowledge through numbers
(숫자를 통한 자기 이해)
Based upon speech patterns, the particular words they used, and even
details as seemingly trivial as whether they said “um” or “err” – and then
utilise these insights to put them through to the agent best suited for dealing
with their emotional needs?
(Chicago’s Mattersight Corporation does exactly that. Based on custom
algorithms, Mattersight calls its business “predictive behavioral routing”.)

Based upon speech patterns, the particular words they used, and even details as seemingly trivial as whether they said “um” or
“err” – and then utilise these insights to put them through to the agent best suited for dealing with their emotional needs?
(Chicago’s Mattersight Corporation does exactly that. Based on custom algorithms, Mattersight calls its business “predictive
behavioral routing”.)
The man behind Mattersight’s behavioural models is a clinical psychologist
named Dr Taibi Kahler. Kahler is the creator of a type of psychological
behavioural profiling called Process Communication.
What Kahler noticed was that certain predictable signs precede particular
incidents of distress, and that these distress signs are linked to specific
speech patterns. These, in turn, led to him developing profiles on the six
different personality types he saw recurring.

Based upon speech patterns, the particular words they used, and even details as seemingly trivial as whether they said “um” or
“err” – and then utilise these insights to put them through to the agent best suited for dealing with their emotional needs?
(Chicago’s Mattersight Corporation does exactly that. Based on custom algorithms, Mattersight calls its business “predictive
behavioral routing”.)
The man behind Mattersight’s behavioural models is a clinical psychologist named Dr Taibi Kahler. Kahler is the creator of a
type of psychological behavioural profiling called Process Communication.
What Kahler noticed was that certain predictable signs precede particular incidents of distress, and that these distress signs are
linked to specific speech patterns. These, in turn, led to him developing profiles on the six different personality types he saw
recurring.
A person patched through to an individual with a similar personality type to
their own will have an average conversation length of five minutes, with a 92
percent problem-resolution rate. A caller paired up to a conflicting
personality type, on the other hand, will see their call length double to ten
minutes – while the problem-resolution rate tumbles to 47 percent.

Personality type Personality traits
How
common?
“Thinkers”
Thinkers view the world through data. Their primary way of dealing with
situations is based upon logical analysis of a situation. They have the potential
to become humourless and controlling.
1 in 4
people
“Rebels”
Rebels interact with the world based on reactions. They either love things or
hate them. Many innovators come from this group. Under pressure they can
be negative and blameful.
1 in 5
people
“Persisters”
Persisters filter everything through their opinions. Everything is measured up
against their world view. This describes the majority of politicians.
1 in 10
people
“Harmonisers”
Harmonisers deal with everything in terms of emotions and relationships.
Tight situations make this group overreactive.
3 in 10
people
“Promoters”
Promoters view everything through action. These are the salesmen of the
world, always looking to close a deal. They can be irrational and impulsive.
1 in 20
people
“Imaginers”
Imaginers deal in unfocused thought and reflection. These people operate in
vivid internal worlds and are likely to spot patterns where others cannot.
1 in 10
people
Dr Taibi Kahler’s the six different personality types

착각과 통찰
• Computers Are Social Actors & Similarity-attraction
 그냥 입력 데이터 분석해서 적정하게 검색 결과를 대화형으로 보여주면 되는 거
아닌가? (X)
 Social Actor로 정의하자! 즉 성격을 부과하자! 특히 초기에 고정(세팅)된 성격
제공(또는 선택)이 아니라 사용자의 특성을 점차 학습하여 사용자와 유사한 성격
유형으로 적응하는 방향을 추구해야 함. 이것이 최근 부각되는 N=1의 사용자 경험
전략 방향임
 이것이 가능하기 위해서는 사용자의 성격 특성을 측정하고 적응할 수 있는
알고리즘 개발이 필요함
 지속적인 구매 등의 지속적인 행동 변화를 도모하는데 큰 기여를 할 수 있음
42

Empathize With Emotions: Humor
• I’ve even heard that comedy writers are becoming the next hot UX hires in the hopes
of making copy more engaging.
• We often use humor to lighten up a situation and smooth over our mistakes.
We might consider using this tactic when the bot encounters a limitation or needs to give
an error message.
• To succeed with humor, consider what type of personality your users would find
engaging.
Make sure to fit the tone to the task at hand, as well as the brand your bot is representing.
For example, if you’re building a banking bot, avoid flippantly humorous error messages
when a user’s balance can’t be retrieved.
This is a very important place to check in with our User Personas to get a feel for their
preferences and their intents.
43
References:
• https://replika.ai/

Empathize With Emotions: Use emojis
• We did learn to only use emojis in positive affirmation responses and to introduce
them later in the onboarding process.
• We use templated responses for many interactions that the user inputs, and one thing
we learned is that people don’t like being “forced” to send emojis back to a bot. Our
responses tend to be fairly generic because people need to connect to the things we
are allowing them to say.
44
References:

Natural Flow and Cadence: First Impression
• First impressions really do make or break an interaction, so we should consider our
greeting carefully.
It’s often helpful to explain what our bot can do in the first interaction,
But it can be jarring for the user to get a huge chunk of text before exchanging a greeting.
Like it or not, your first-time users are essentially talking with a stranger, which in the
range of enjoyable life activities ranks right up there with surprise dental emergencies.
They may grow to love your bot at some point, but in the first engagement, your bot is
essentially the guy on the subway who just started asking them questions.
45
References:
• http://venturebeat.com/2016/10/05/chatbot-design-how-to-build-a-beautiful-ui-when-all-you-have-is-text/

Natural Flow and Cadence: Don’t Be A Broken Record
You: What is your name?
Mitsuku: My name is Mitsuku.
You: What is your name?
Mitsuku: You just asked me that. It’s still Mitsuku. It hasn’t changed in the last 10
secs.
• Nobody likes being told the same thing over and over again,
So why do chatbots keep doing it? Bots should detect when they’re about to repeat a
previously given answer and switch strategies.
As an egregious example, I ran into a bug with the 1–800-Flowers chatbot on Facebook
Messenger. This chatbot presents categories and subcategories of bouquets you can choose
from, but if you go into a subcategory card menu, the previous menus become deactivated
and there’s no obvious way to go back to them. Pretty silly, eh?
46
References:
• http://designrfix.com/design/key-principles-chatbot-design

Writing scripts
• Our audience comes from a huge cross-
section of society. We sell our product
directly to employers of a highly educated
workforce as well as to people who didn’t
finish high school. We’ve found that,
regardless of the audience, making your
scripts as simple as possible results in the
greatest level of engagement.
• You can use the Flesch-Kincaid Grade Level
feature in Microsoft Word or an app like
Hemingwayapp.com to figure out the
present grade level of your scripts. I don’t
think there is a hard and fast rule of what
grade level you want to write to, but in
general, the lower the better.
47
References:
• http://beta.hemingwayapp.com/

New Command Line Interface(CLI)? Prompt chatbot
• A Prompt bot in Slack
Typing "@uber from Penn Station to LaGuardia Airport" would order a car to pick you up
to drop you at the airport.
Typing "@nest 73" and Prompt will turn up your heat if you have a Nest thermostat.
Typing "@flightstats AA21" would get you the status for American Airlines flight 21
Typing "@hue" will turn on and off your Philips Hue bulbs
Typing "@showtimes" will tell you what movies are playing near you, and so on.
• Just like the command line, though, Prompt's biggest barrier to entry is syntax.
That's especially frustrating on smartphones, where on-screen keyboards and
autocorrect can wreak havoc with Prompt's super-specific vocabulary.
49
References:
• https://www.fastcodesign.com/3057596/prompt-wants-to-be-the-one-chatbot-to-rule-them-all
• http://www.message.io/

Watson for Thermocare; 비접촉성 체온계; 대화형 모바일 헬스케어 시스템 지향
50
Image Source: https://youtu.be/a6BUoUeH4OY

Watson for Thermocare(cont.)
51
Image Source: https://youtu.be/a6BUoUeH4OY
(아이 등 환자의)
개인 정보(특정 질병 등)에
기반한 응급 조치 권고와
시스템 자동 조치(진료 예약,
119 등) 필요
(부모 등 구매자의)
개인별 성향에 기반하고
권위와 신뢰(의사 등)을 투영한
환경(응급 등)에 적합한
(부모 등 구매자의) 심리 배려 및
(아이 등 환자 대상의) 실효적인 솔루션
제공 목적의 대화형 시스템 접근 필요
체온계와 모바일 디바이스의 역할 분담
고려 필요; 차후 음성 대화 시스템
탑재(체온계; 아마존 에코 등) 및 상세한
정보 접근(모바일 디바이스)

Timeline Display and Screenshot Edit in a Chat Room by Xiaoxue(Ellie) Zhang
52
References:
• https://medium.muz.li/amazing-chat-interface-inspiration-9ce35222b93a#.96gv6zl7w
• https://dribbble.com/xiaoxuezhang

Messenger Bots & Partial payment Concept by Isil Uzum
53
References:
• https://medium.muz.li/amazing-chat-interface-inspiration-9ce35222b93a#.96gv6zl7w
• https://dribbble.com/isiluzum

경청해주셔서
고맙습니다!

designing conversations: Conversational interfaces, Bot Interactions, Chatbot as personalities (UX 전략적 통찰과 방향성 중심으로)

More Related Content

What's hot (20)

Viewers also liked (8)

Similar to designing conversations: Conversational interfaces, Bot Interactions, Chatbot as personalities (UX 전략적 통찰과 방향성 중심으로) (20)

More from Billy Choi (20)

Recently uploaded (20)

designing conversations: Conversational interfaces, Bot Interactions, Chatbot as personalities (UX 전략적 통찰과 방향성 중심으로)