TTI / Vanguard Understanding Understanding Conference Pittsburgh 3 October 2012
Understanding Relevance Hugh Dubberly Dubberly Design Office
Identity for a person
Relevance The right resources in the right amounts at the right time in the right place
or team
=
Context
+
to accomplish the tasks at hand without disruption or loss of flow
Dubberly Design Office · Understanding Relevance · 3 October 2012
2
Understanding relevance is already a big business.
Dubberly Design Office · Understanding Relevance · 3 October 2012
3
A successful bid for the search term “is auto insurance required” through Google Adwords can cost $143 per click – that’s 1 click.
Dubberly Design Office · Understanding Relevance · 3 October 2012
4
Our models of relevance are still pretty crude even at the leaders – Amazon, Facebook, and Google.
Dubberly Design Office · Understanding Relevance · 3 October 2012
5
Understanding relevance promises a new era of CRM. (customer relationship management)
Relevance transforms advertising from interruption to something more like assistance.
Fading
Today’s standard
Emerging
Mass
Personalized
Conversation
Sift through broadcast streams
Search the web for things you want
Interesting things find you
Dubberly Design Office · Understanding Relevance · 3 October 2012
6
Understanding relevance promises a new era of UX. (user experience)
Relevance expands UX from simple interaction to something more like conversation.
Fading
Today’s standard
Emerging
Command line
Direct manipulation
Delegate to agents
Remember + type
See + point
Ask + tell
Dubberly Design Office · Understanding Relevance · 3 October 2012
7
Better understanding of relevance is coming from several simultaneous revolutions.
Sensors
Networks
Combinatorial Innovation
Cloud-based Computing Dubberly Design Office · Understanding Relevance · 3 October 2012
Mobile Computers 8
Sensors will be ubiquitous – at checkpoints – logging everything you do online – all around you – on you Identity – in you
Context
Relevance Community
Goals
Conversati
on
You
Data Tracking + Analysis
Measures
Actions
Effects
Sensors
Dubberly Design Office · Understanding Relevance · 3 October 2012
Information + Tools + Coaching
Disturbances
9
Sensors are being printed – like micro-processor chips; quantities are increasing; prices are dropping.
Understanding Relevance
10
Sensors are connecting – forming mesh networks. Each vine has a sensor; each sensor talks to the next; hubs connect to the internet, providing a heat and humidity map.
Dubberly Design Office · Understanding Relevance · 3 October 2012
11
The same story is playing out in healthcare; with many companies offering home network hubs, connecting medical diagnostic devices to the internet.
Alere
Philips
Philips
Intel
DLM212
Telestation
Motiva (displayed through TV)
Health Guide PHS6000
Care Innovations
Care Innovations
Care Innovations
GrandCare Systems
Guide Tablet
Connect
QuietCare sensors and data communicator
Grandcare Interactive
Dubberly Design Office · Understanding Relevance · 3 October 2012
12
Healthcare hubs are one of many types of home network hubs; for now the market is expanding, but standards are emerging, and network effects mean consolidation is inevitable. Security
Media
Appliance/Energy
Routers
ADT
Logitech
GE
Apple
Pulse Home Security
Harmony 1100 Advanced Universal Remote
Nucleus Energy Manager
Airport Extreme
GE
Control4
Nest
Cisco
FrontPoint Security Touchscreen
7” In-Wall Touch Screen with Camera
The Learning Thermostat
Linksys EA4500
Dubberly Design Office · Understanding Relevance · 3 October 2012
13
Expanding networks deliver rapidly growing data streams for processing by massive cloud-based computer systems which deliver the results almost anywhere.
Sensors Dubberly Design Office · Understanding Relevance · 3 October 2012
Cloud-based Computing
Mobile Devices 14
What does this mean? Massive data collecting – organized into a taxonomy of personal identity Income + Expenses Liquid Assets Possessions Resource Use Insurance
Schools + Courses Employees + Jobs Skills + Expertise
Work Finance
Interests
Civic Voting + Registration Juries + Offices Licenses, Visas, Passports Taxes, Fees, Finances
Wellbeing
Grooming + Prevention Body Systems Emotions + Affect Exercises + Diet
Social
Family + Friends Roommates + Neighbors Colleagues + Acquaintances Dubberly Design Office · Understanding Relevance · 3 October 2012
Hobbies Create/Participate Media Curate/Coach × Performances Comment Sports Consume
Meetings Conversations × Agreements Transactions 15
Drilling into a sub-categories shows the potential for detail;
today lab tests can measure over 150 analytes; more are in development.
Pituitary gland
(hydrocortisone)
Normal, PM: 3–17 µg/dL
17 Hydroxyprogesterone * See also: ovaries
Man, normal: .06–3 mg/L Woman (follicular phase), normal: .2–1 mg/L
Angiotensinconverting enzyme (ACE)
Normal: 23–57 U/L
Growth hormone
At peak: 5–45 ng/mL Between peaks: < 5 ng/mL
Follicle-stimulating hormone (FSH)
Prepubertal: < 1 – 3 IU/L Adult male: 1–8 IU/L Adult female (follicular & luteal phase): 1–11 IU/L Adult female (ovulation): 6–26 IU/L Post-menopausal female: 30–118 IU/L
Adrenocorticotropic hormone (ACTH)
Normal: 20–80 pg/mL
Prolactin
Female, normal: < 20 ng/mL Male, normal: < 15 ng/mL
Blood Glucose
Hypoglycemia: < 3 mmol/l Normal: 3.6–5.8 mmol/l Normal, post-meal: 7 mmol/l (chronicly)
Luteinizing hormone (LH)
Female (peak): 20–75 IU/L emale (post-menopausal): 15–60 IU/L
Insulin absorption
-
Plasma osmolality
Normal: 275–295 mOsm/kg
Work
ce
Interests Grooming + Prevention
c
Wellbeing Social
Body Systems
Cardiovascular
Blood
Emotions + Affect Exercises + Diet
Dubberly Design Office · Understanding Relevance · 3 October 2012
16
Identity changes over time.
Sc
Se
rvi
ce
s/A
cc
ou
Dubberly Design Office · Understanding Relevance · 3 October 2012
ore
nts
/Cr
Act
s
ed
en
tia
ivit
ies
ls
17
Activities take place within contexts. Physical Contexts – Who (actors) – What (resources) – Where (location) – When (time) Why How
Goals Tasks
Cultural Contexts (task domains) – Languages (vocabulary) – Concepts (knowledge) – Methods (skills) – Rules (norms)
Recent Dubberly Design Office · Understanding Relevance · 3 October 2012
Current
Anticipated 18
The space of possible services is sparsely populated.
Goals/Tasks/Queues – Opportunities – Risks
Civic
Finance
Kickstarter
Motif
Status / Alerts – Strengths – Weaknesses Records – Successes – Failures
Dubberly Design Office · Understanding Relevance · 3 October 2012
Work
Interests
Wellbeing
Netflix Queue Amazon Wish List Air BNB
Online Wallets Square
Mint
Social
Facebook
Linkedin Resume
Evernote Amazon Purchases
EMR/PHR
19
Doc Searls predicts a new order in which you control your data; CRM gives way to VRM and you control your vendor relationships. “In the not-too-distant future, you will be able, for example, to change your contact information with many vendors at once, rather than many times, over and over, at many different websites. You will declare your own policies, preferences and terms of engagement—and do it in ways that can be automated both for you and the companies you engage.”
—”The Customer As A God,” WSJ, July 21, 2012 Dubberly Design Office · Understanding Relevance · 3 October 2012
20
That’s the emerging future, but where are we today?
Dubberly Design Office · Understanding Relevance · 3 October 2012
21
Networked services can recognize their users and respond uniquely. recognized by
user
networked service
receives unique response from
Dubberly Design Office · Understanding Relevance · 3 October 2012
22
Networked services collect information as a natural part of operating. collect
recognized by
user
networked service
data
receives unique response from
Dubberly Design Office · Understanding Relevance · 3 October 2012
23
Networked services change continuously.
collect
recognized by
user
networked service
informs
data
designer
receives unique response from continuously revised by
Dubberly Design Office · Understanding Relevance · 3 October 2012
24
“... designing networked services requires a new way of thinking about a product and its development.” —Tim Misner networked service hardware
designer A
designer A
hardware
designer A
Dubberly Design Office · Understanding Relevance · 3 October 2012
networked data networkeddata service service
data
hardware
vs
vs
vs
designer B
designer B
designer B
25
“...internal discussion changes from
‘what features or quality level do we think our products need?’
to
‘what data can we collect about our features and quality?’”
Dubberly Design Office · Understanding Relevance · 3 October 2012
26
With traditional hardware products, designers have limited knowledge of customer use patterns.
makes first version of
is used by
may provide feedback through
may suggest insights to
makes next version of designer/ manufacturer
hardware
Dubberly Design Office · Understanding Relevance · 3 October 2012
customer
sampling techniques - customer support call - survey - usability study - focus group - ethnographic study
27
With web-based services, designers can have almost complete knowledge of customers behavior.
may provide feedback through
is used by
makes first version of
actions recorded in
sampling techniques
may suggest insights to
provide hard data for decisions
makes next version of designer/ publisher
website
Dubberly Design Office · Understanding Relevance · 3 October 2012
customer
website logs + analytics tools
28
As hardware products become part of networked services, they become more like web sites.
may provide feedback through
is used by
makes first version of
actions recorded in
sampling techniques
may suggest insights to
provide hard data for decisions
sends updated software to designer/ manufacturer
hardware
Dubberly Design Office · Understanding Relevance · 3 October 2012
customer
website logs + analytics tools
29
Special thanks to Paul Pangaro Michael Liebhold Michael Gallagher
[email protected] Presentation posted at www.dubberly.com/presentations/Understanding_Relevance.pdf
Appendix
Dubberly Design Office · Understanding Relevance · 3 October 2012
31
input
Dubberly Design Office · Understanding Relevance · 3 October 2012
process
output
32
goal
measure
compare
act
environment
disturbance
Dubberly Design Office · Understanding Relevance · 3 October 2012
33
goal
measure
compare
act
(set) goal
measure
compare
act
environment
disturbance
Dubberly Design Office · Understanding Relevance · 3 October 2012
34
are explained by
models and stories are tools for thinking
models
stories
models and stories are tools for discussion
create
Dubberly Design Office · Understanding Relevance · 3 October 2012
35
Human communication relies on agreement.
Do we seem to agree, that we agree?
my model of the correspondence of your model of the subject to my model of the subject (Do we seem to agree?)
my model of your model of the subject
my model of the subject
your model of the subject
me
you
subject
Dubberly Design Office · Understanding Relevance · 3 October 2012
36
Conversations may lead to trust. Trust Relationship (Trans)action Agreement Understanding
Dubberly Design Office · Understanding Relevance · 3 October 2012
37
This model describes the learning process.
SECI model of knowledge creation Ikujiro Nonaka (1995) Explicit 1
Combination Connecting
Externalization Articulating
SECI Model of Knowledge Creation
Systemizing and applying explicit knowledge and information 7. Gathering and integrating explicit knowledge 8. Transferring and diffusing explicit knowledge 9. Editing explicit knowledge
Articulating tacit knowledge through dialogue and reflection 5. Articulating tacit knowledge 6. Translating tacit knowledge
Internalization Embodying
Socialization Empathizing Sharing and creating tacit knowledge through direct experience 1. Walking around inside the company 2. Walking around outside the company 3. Accumulating tacit knowledge 4. Transferring tacit knowledge
Individual Group
Explicit 2
Tacit 2
Ikujiro Nonaka (1995)
Learning and acquiring new tacit knowledge in practice 10. Embodying explicit knowledge through action and practice 11. Using simulation and experiments
Tacit 1 Tacit 3
Organization Community of organizations Knowledge conversion spiral
Dubberly Design Office · Understanding Relevance · 3 October 2012
38
Designing is analogous to learning. SECI model of knowledge creation Ikujiro Nonaka (1995)
Analysis-Synthesis Bridge Model Dubberly, Evenson & Robinson (2008) Abstract
suggest
What “could be”
Preferred – Explicit (Future)
distilled
Internalization Embodying
Socialization Empathizing Sharing and creating tacit knowledge through direct experience 1. Walking around inside the company 2. Walking around outside the company 3. Accumulating tacit knowledge 4. Transferring tacit knowledge
Individual Group
Explicit 2
Existing – Implicit (Current)
t as
What “is”
manifes
Concrete
Systemizing and applying explicit knowledge and information 7. Gathering and integrating explicit knowledge 8. Transferring and diffusing explicit knowledge 9. Editing explicit knowledge
Articulating tacit knowledge through dialogue and reflection 5. Articulating tacit knowledge 6. Translating tacit knowledge
Model of what “could be”
to
Model of what “is”
Describe
Combination Connecting
Externalization Articulating
Tacit 2
Interpret
Explicit 1
Prototyping
Researching
Learning and acquiring new tacit knowledge in practice 10. Embodying explicit knowledge through action and practice 11. Using simulation and experiments
Tacit 1 Tacit 3
Organization Community of organizations Knowledge conversion spiral
Analysis-Synthesis Bridge Model
SECI Model of Knowledge Creation
Dubberly, Evenson & Robison (2008)
Ikujiro Nonaka (1995)
Dubberly Design Office · Understanding Relevance · 3 October 2012
39