Upstream versus Downstream interventions
Banny Banerjee
Associate Professor
Stanford d.school, Stanford Design Program
Director, Stanford ChangeLabs
Associate Professor
d.
Mechanical Engineering Department
Stanford University
With contributions from
Jason Bade
Johannes Zachrisson Daae
David Miller
S TA N F O R D
DESIGN
ChangeLabs.
PROGRAM
Global Score Card
2030 Targets
Confidence Level
50% reductions in CO2e
45-50% more FOOD
39-45% more ENERGY
40% more FRESHWATER
3-4 orders of magnitude
LOW
slowdown in EXTINCTION rates
0
in EXTREME POVERTY
($1.25/Day)
LOW
LOW
LOW
NIL
LOW
Sources: The energy challenge, Mike Hightower1 & Suzanne A. Pierce, Nature 452, 285-286 (20 March 2008). Word Bank 2013 Water report, Information brief on Water and Agriculture
in the Green Economy. UNW-DPAC, 2011, Oct 9, 2013 World Bank announcement on Poverty Eradication Goals
Banny Banerjee, Stanford University
Banny Banerjee, Stanford University
Banny Banerjee, Stanford University
intervention
challenge
Banny Banerjee, Stanford University
“What kind of interventions will change the ecosystem behavior?”
Banny Banerjee, Stanford University
In all it’s messiness
Banny Banerjee, Stanford University
OVER-ARCHING QUESTIONS AND ISSUES
Banny Banerjee, Stanford University
OVER-ARCHING QUESTIONS AND ISSUES
How do we make SIGNIFICANT reductions in energy usage and GHGs?
Banny Banerjee, Stanford University
OVER-ARCHING QUESTIONS AND ISSUES
How do we make SIGNIFICANT reductions in energy usage and GHGs?
How do we make it financially viable? Revenue positive?
Banny Banerjee, Stanford University
OVER-ARCHING QUESTIONS AND ISSUES
How do we make SIGNIFICANT reductions in energy usage and GHGs?
How do we make it financially viable? Revenue positive?
How do we create a business ecosystem when a KWH is so cheap?
Banny Banerjee, Stanford University
OVER-ARCHING QUESTIONS AND ISSUES
How
How
How
How
do we make SIGNIFICANT reductions in energy usage and GHGs?
do we make it financially viable? Revenue positive?
do we create a business ecosystem when a KWH is so cheap?
to get the plans past our internal organizational and decision systems?
Banny Banerjee, Stanford University
OVER-ARCHING QUESTIONS AND ISSUES
How
How
How
How
How
do we make SIGNIFICANT reductions in energy usage and GHGs?
do we make it financially viable? Revenue positive?
do we create a business ecosystem when a KWH is so cheap?
to get the plans past our internal organizational and decision systems?
do we diffuse the enabling technologies at scale?
Banny Banerjee, Stanford University
OVER-ARCHING QUESTIONS AND ISSUES
How do we make SIGNIFICANT reductions in energy usage and GHGs?
How do we make it financially viable? Revenue positive?
How do we create a business ecosystem when a KWH is so cheap?
How to get the plans past our internal organizational and decision systems?
How do we diffuse the enabling technologies at scale?
Getting people to save energy is harder than getting them to floss their teeth!
Banny Banerjee, Stanford University
OVER-ARCHING QUESTIONS AND ISSUES
How do we make SIGNIFICANT reductions in energy usage and GHGs?
How do we make it financially viable? Revenue positive?
How do we create a business ecosystem when a KWH is so cheap?
How to get the plans past our internal organizational and decision systems?
How do we diffuse the enabling technologies at scale?
Getting people to save energy is harder than getting them to floss their teeth!
Death by a thousand cuts
Banny Banerjee, Stanford University
OVER-ARCHING QUESTIONS AND ISSUES
How do we make SIGNIFICANT reductions in energy usage and GHGs?
How do we make it financially viable? Revenue positive?
How do we create a business ecosystem when a KWH is so cheap?
How to get the plans past our internal organizational and decision systems?
How do we diffuse the enabling technologies at scale?
Getting people to save energy is harder than getting them to floss their teeth!
Death by a thousand cuts
When you want a toast, you want a toast!
Banny Banerjee, Stanford University
OVER-ARCHING QUESTIONS AND ISSUES
How do we make SIGNIFICANT reductions in energy usage and GHGs?
How do we make it financially viable? Revenue positive?
How do we create a business ecosystem when a KWH is so cheap?
How to get the plans past our internal organizational and decision systems?
How do we diffuse the enabling technologies at scale?
Getting people to save energy is harder than getting them to floss their teeth!
Death by a thousand cuts
When you want a toast, you want a toast!
Okay to spend $2000 on a bigger TV but not on weatherizing one’s home
Banny Banerjee, Stanford University
OVER-ARCHING QUESTIONS AND ISSUES
How do we make SIGNIFICANT reductions in energy usage and GHGs?
How do we make it financially viable? Revenue positive?
How do we create a business ecosystem when a KWH is so cheap?
How to get the plans past our internal organizational and decision systems?
How do we diffuse the enabling technologies at scale?
Getting people to save energy is harder than getting them to floss their teeth!
Death by a thousand cuts
When you want a toast, you want a toast!
Okay to spend $2000 on a bigger TV but not on weatherizing one’s home
Mental models - recycling a soda can before boarding a flight
Banny Banerjee, Stanford University
Banny Banerjee, Stanford University
ENABLING
TECHNOLOGIES
Banny Banerjee, Stanford University
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
Banny Banerjee, Stanford University
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
SIN
MO ESS
DE
L
Banny Banerjee, Stanford University
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
SIN
MO ESS
DE
L
ORG BEHAVIOR
& DECISION
SYSTEMS
Banny Banerjee, Stanford University
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
SIN
MO ESS
DE
L
ORG BEHAVIOR
& DECISION
SYSTEMS
&
Y
LIC RY
O
P ATO TE
UL IMA
G
CL
RE
Banny Banerjee, Stanford University
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
SIN
MO ESS
DE
L
ORG BEHAVIOR
& DECISION
SYSTEMS
IN
VE
S
TM
EN
T
&
Y
LIC RY
O
P ATO TE
UL IMA
G
CL
RE
Banny Banerjee, Stanford University
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
SIN
MO ESS
DE
L
ORG BEHAVIOR
& DECISION
SYSTEMS
IN
VE
S
TM
EN
T
&
Y
LIC RY
O
P ATO TE
UL IMA
G
CL
RE
GOAL
Banny Banerjee, Stanford University
TH
CH EORI
AN ES
GE OF
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
SIN
MO ESS
DE
L
ORG BEHAVIOR
& DECISION
SYSTEMS
IN
VE
S
TM
EN
T
&
Y
LIC RY
O
P ATO TE
UL IMA
G
CL
RE
GOAL
Banny Banerjee, Stanford University
SIN
MO ESS
DE
L
TH
CH EORI
AN ES
GE OF
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
G
IN
D
AN R
T
RS VIO
E
D
A
UN BEH
OF
ORG BEHAVIOR
& DECISION
SYSTEMS
IN
VE
S
TM
EN
T
&
Y
LIC RY
O
P ATO TE
UL IMA
G
CL
RE
GOAL
Banny Banerjee, Stanford University
SIN
MO ESS
DE
L
TH
CH EORI
AN ES
GE OF
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
G
IN
D
AN R
T
RS VIO
E
D
A
UN BEH
OF
ORG BEHAVIOR
& DECISION
SYSTEMS
&
Y
LIC RY
O
P ATO TE
UL IMA
G
CL
RE
IN
VE
S
TM
EN
T
MO
TIV
AT
IO
NS
GOAL
Banny Banerjee, Stanford University
TH
CH EORI
AN ES
GE OF
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
SIN
MO ESS
DE
L
G
IN
D
AN R
T
RS VIO
E
D
A
UN BEH
OF
ORG BEHAVIOR
& DECISION
SYSTEMS
&
Y
LIC RY
O
P ATO TE
UL IMA
G
CL
RE
TM
VE
S
IO
NS
S
IN
AT
RM
GOAL
TIV
NO
EN
T
MO
Banny Banerjee, Stanford University
TH
CH EORI
AN ES
GE OF
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
SIN
MO ESS
DE
L
ORG BEHAVIOR
& DECISION
SYSTEMS
G
IN
D
AN R
T
RS VIO
E
D
A
UN BEH
OF
INNOVATION
PROCESS
&
Y
LIC RY
O
P ATO TE
UL IMA
G
CL
RE
TM
VE
S
IO
NS
S
IN
AT
RM
GOAL
TIV
NO
EN
T
MO
Banny Banerjee, Stanford University
TH
CH EORI
AN ES
GE OF
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
SIN
MO ESS
DE
L
ORG BEHAVIOR
& DECISION
SYSTEMS
G
IN
D
AN R
T
RS VIO
E
D
A
UN BEH
OF
INNOVATION
PROCESS
&
Y
LIC RY
O
P ATO TE
UL IMA
G
CL
RE
TM
VE
S
IO
NS
S
IN
AT
RM
GOAL
TIV
NO
EN
T
MO
Banny Banerjee, Stanford University
TH
CH EORI
AN ES
GE OF
ENABLING
TECHNOLOGIES
G
IN Y
AL
SC TEG
RA
ST
BU
SIN
MO ESS
DE
L
G
IN
D
AN R
T
RS VIO
E
D
A
UN BEH
OF
OR
ORG BEHAVIOR
& DECISION
SYSTEMS
INNOVATION
PROCESS
&
Y
LIC RY
O
P ATO TE
UL IMA
G
CL
RE
TM
VE
S
IO
NS
S
IN
AT
RM
GOAL
TIV
NO
EN
T
MO
Banny Banerjee, Stanford University
Behavior
Banny Banerjee, Stanford University
The cheapest barrel of oil is the one that we don’t use
Banny Banerjee, Stanford University
Banny Banerjee, Stanford University
For most systems, behavior is a choke point
Banny Banerjee, Stanford University
For most systems, behavior is a choke point
Behavior change allows demand-side interventions
Banny Banerjee, Stanford University
For most systems, behavior is a choke point
Behavior change allows demand-side interventions
Behavior change can scale faster than infrastructure
Banny Banerjee, Stanford University
CHALLENGES AND OPPORTUNITIES!
IN AN ALTERED ECOSYSTEM!
CLOUD!
PV!
Sensors!
Feedback, !
Nudges,!
Incentives!
Mobile!
Device!
Smart !
Controls!
E
V
Appliances!
Storage!
Smart Meter!
( + Smart Grid)!
Changing regulations, pricing structures, markets, competitive landscape, energy futures, sensibilities!
tariffs, social norms, relationship with home-owners, and information landscapes!
Banny Banerjee, Stanford University
Upstream
Interventions
Downstream
Interventions
BannyBanerjee,
Banerjee,Stanford
StanfordUniversity
University
Banny
How do we change the energy
consumption of a region?
How do we get people to use a
different thermostat set-point
BannyBanerjee,
Banerjee,Stanford
StanfordUniversity
University
Banny
US Home Energy use breakdown
#
)3
&
*3
"
*3
,3
'
-3
(Gardner & Stern)
.3
,,3
#&$
-3
/3
03
**3
**3
*+3
Banny Banerjee, Stanford University
Upstream
Interventions
Downstream
Interventions
Target Profile
Contexts
Banny Banerjee, Stanford University
KWH
CAPITA
ENERGY
SOURCE
INFRASTRUCTURE
BEHAVIOR
PATTERNS
KWH
CAPITA
PEAK
KW
CURRENT
ECOLOGY
MENTAL
MODELS,
FRAMES
STANDARDS
BUSINESS
MODELS
PV/
ALT-SOURCE
INFRASTRUCTURE
PEAK
KW
FUTURE
ECOLOGY
STANDARDS
BIZ
MODELS
BEHAVIOR
FRAMES
Banny Banerjee, Stanford University
KWH
CAPITA
ENERGY
SOURCE
INFRASTRUCTURE
BEHAVIOR
PATTERNS
KWH
CAPITA
PV/
ALT-SOURCE
PEAK
KW
CURRENT
ECOLOGY
MENTAL
MODELS,
FRAMES
STANDARDS
BUSINESS
MODELS
f
INFRASTRUCTURE
PEAK
KW
FUTURE
ECOLOGY
STANDARDS
BIZ
MODELS
BEHAVIOR
FRAMES
Banny Banerjee, Stanford University
Scaled Innovation
KWH
CAPITA
ENERGY
SOURCE
INFRASTRUCTURE
BEHAVIOR
PATTERNS
KWH
CAPITA
PEAK
KW
CURRENT
ECOLOGY
MENTAL
MODELS,
FRAMES
STANDARDS
BUSINESS
MODELS
PV/
ALT-SOURCE
INFRASTRUCTURE
PEAK
KW
FUTURE
ECOLOGY
STANDARDS
BIZ
MODELS
BEHAVIOR
FRAMES
Banny Banerjee, Stanford University
Scaled Innovation
Behavioral, Technical &
Business Determinants
KWH
CAPITA
ENERGY
SOURCE
INFRASTRUCTURE
BEHAVIOR
PATTERNS
KWH
CAPITA
PEAK
KW
CURRENT
ECOLOGY
MENTAL
MODELS,
FRAMES
STANDARDS
BUSINESS
MODELS
PV/
ALT-SOURCE
INFRASTRUCTURE
PEAK
KW
FUTURE
ECOLOGY
STANDARDS
BIZ
MODELS
BEHAVIOR
FRAMES
Banny Banerjee, Stanford University
Scaled Innovation
Behavioral, Technical &
Business Determinants
KWH
CAPITA
ENERGY
SOURCE
INFRASTRUCTURE
BEHAVIOR
PATTERNS
KWH
CAPITA
PV/
ALT-SOURCE
PEAK
KW
CURRENT
ECOLOGY
MENTAL
MODELS,
FRAMES
STANDARDS
BUSINESS
MODELS
Strategic Framing
Scaled Concepts
INFRASTRUCTURE
PEAK
KW
FUTURE
ECOLOGY
STANDARDS
BIZ
MODELS
BEHAVIOR
FRAMES
Banny Banerjee, Stanford University
Scaled Innovation
Behavioral, Technical &
Business Determinants
KWH
CAPITA
ENERGY
SOURCE
INFRASTRUCTURE
BEHAVIOR
PATTERNS
KWH
CAPITA
PV/
ALT-SOURCE
PEAK
KW
CURRENT
ECOLOGY
MENTAL
MODELS,
FRAMES
STANDARDS
Strategic Framing
Scaled Concepts
BUSINESS
MODELS
INFRASTRUCTURE
PEAK
KW
FUTURE
ECOLOGY
BIZ
MODELS
BEHAVIOR
Strategic Roadmaps
Implementation & Scaling
STANDARDS
FRAMES
Banny Banerjee, Stanford University
Scaled Innovation
Behavioral, Technical &
Business Determinants
Strategic Framing
Scaled Concepts
Strategic Roadmaps
Implementation & Scaling
Banny Banerjee, Stanford University
Scaled Innovation
Behavioral, Technical &
Business Determinants
Behavioral, Technological DRIVERS
Extrinsic factors: SITUATIONAL CONSTRAINTS
Ecosystem Analysis: STAKEHOLDER RELATIONSHIPS
Strategic Framing
Scaled Concepts
Strategic Roadmaps
Implementation & Scaling
Banny Banerjee, Stanford University
Scaled Innovation
Behavioral, Technical &
Business Determinants
Behavioral, Technological DRIVERS
Extrinsic factors: SITUATIONAL CONSTRAINTS
Ecosystem Analysis: STAKEHOLDER RELATIONSHIPS
STRATEGIC PATHWAYS & PLATFORMS
Strategic Framing
Scaled Concepts
Design of SOLUTIONS, SERVICES, BIZ MODELS
SCALING and DIFFUSION STRATEGY
Strategic Roadmaps
Implementation & Scaling
Banny Banerjee, Stanford University
Scaled Innovation
Behavioral, Technical &
Business Determinants
Behavioral, Technological DRIVERS
Extrinsic factors: SITUATIONAL CONSTRAINTS
Ecosystem Analysis: STAKEHOLDER RELATIONSHIPS
STRATEGIC PATHWAYS & PLATFORMS
Strategic Framing
Scaled Concepts
Design of SOLUTIONS, SERVICES, BIZ MODELS
SCALING and DIFFUSION STRATEGY
Strategic Roadmaps
Implementation & Scaling
Organizational & Decision Processes
Strategic Partnerships
Rapid Pilots and Implementation rollouts
Banny Banerjee, Stanford University
The innovation process is like an accordion
Banny Banerjee, Stanford University
Cognitive intervention
Piggyback intervention
Stealth intervention
Banny Banerjee, Stanford University
KWH Potential
High
Low
Low
High
Behavior
Banny Banerjee, Stanford University
KWH Potential
High
Low
Low
High
Behavior
Banny Banerjee, Stanford University
KWH Potential
High
Vs
Low
Low
High
Behavior
High Conversion
of Small Segment
Low Conversion
of Larger Segment
Banny Banerjee, Stanford University
Prospect Theory
Losses are felt twice as
much as equivalent gains.
(Kahneman and Tversky 1974)
Banny Banerjee, Stanford University
Hyperbolic Discounting
Banny Banerjee, Stanford University
PRINCIPLES
Banny Banerjee, Stanford University
PRINCIPLES
Understand the local cultural, behavioral and motivational frames
Banny Banerjee, Stanford University
PRINCIPLES
Understand the local cultural, behavioral and motivational frames
Parse the qualitative insights through behavioral lenses
Banny Banerjee, Stanford University
PRINCIPLES
Understand the local cultural, behavioral and motivational frames
Parse the qualitative insights through behavioral lenses
Business models and organizational behavior can ultimately limit success
Banny Banerjee, Stanford University
PRINCIPLES
Understand the local cultural, behavioral and motivational frames
Parse the qualitative insights through behavioral lenses
Business models and organizational behavior can ultimately limit success
Consider hybrid behavioral-techno-business strategies
Banny Banerjee, Stanford University
PRINCIPLES
Understand the local cultural, behavioral and motivational frames
Parse the qualitative insights through behavioral lenses
Business models and organizational behavior can ultimately limit success
Consider hybrid behavioral-techno-business strategies
Upstream platforms that enable multiple downstream interventions
Banny Banerjee, Stanford University
PRINCIPLES
Understand the local cultural, behavioral and motivational frames
Parse the qualitative insights through behavioral lenses
Business models and organizational behavior can ultimately limit success
Consider hybrid behavioral-techno-business strategies
Upstream platforms that enable multiple downstream interventions
When known solutions do not exist, innovate!
Banny Banerjee, Stanford University
PRINCIPLES
Understand the local cultural, behavioral and motivational frames
Parse the qualitative insights through behavioral lenses
Business models and organizational behavior can ultimately limit success
Consider hybrid behavioral-techno-business strategies
Upstream platforms that enable multiple downstream interventions
When known solutions do not exist, innovate!
Allow matches between motivational frames and solution frames
Banny Banerjee, Stanford University
PRINCIPLES
Understand the local cultural, behavioral and motivational frames
Parse the qualitative insights through behavioral lenses
Business models and organizational behavior can ultimately limit success
Consider hybrid behavioral-techno-business strategies
Upstream platforms that enable multiple downstream interventions
When known solutions do not exist, innovate!
Allow matches between motivational frames and solution frames
If you don’t build in scale, it will be harder to build it in later
Banny Banerjee, Stanford University
QUESTIONS
Banny Banerjee, Stanford University
QUESTIONS
Is the segmentation based on motivation, behaviors, and identity?
Banny Banerjee, Stanford University
QUESTIONS
Is the segmentation based on motivation, behaviors, and identity?
Is it a Pull or Push intervention?
Banny Banerjee, Stanford University
QUESTIONS
Is the segmentation based on motivation, behaviors, and identity?
Is it a Pull or Push intervention?
Will it self propel or do you have to scale distribution?
Banny Banerjee, Stanford University
QUESTIONS
Is the segmentation based on motivation, behaviors, and identity?
Is it a Pull or Push intervention?
Will it self propel or do you have to scale distribution?
What is the driving motivation for the user?
Banny Banerjee, Stanford University
QUESTIONS
Is the segmentation based on motivation, behaviors, and identity?
Is it a Pull or Push intervention?
Will it self propel or do you have to scale distribution?
What is the driving motivation for the user?
Is it a Cognitive, Piggyback, or Stealth intervention?
Banny Banerjee, Stanford University
QUESTIONS
Is the segmentation based on motivation, behaviors, and identity?
Is it a Pull or Push intervention?
Will it self propel or do you have to scale distribution?
What is the driving motivation for the user?
Is it a Cognitive, Piggyback, or Stealth intervention?
Is it a Downstream intervention or a Upstream Intervention?
Banny Banerjee, Stanford University
QUESTIONS
Is the segmentation based on motivation, behaviors, and identity?
Is it a Pull or Push intervention?
Will it self propel or do you have to scale distribution?
What is the driving motivation for the user?
Is it a Cognitive, Piggyback, or Stealth intervention?
Is it a Downstream intervention or a Upstream Intervention?
Is it a Single Agency play or Multi-Agency requiring strategic partnerships?
Banny Banerjee, Stanford University
QUESTIONS
Is the segmentation based on motivation, behaviors, and identity?
Is it a Pull or Push intervention?
Will it self propel or do you have to scale distribution?
What is the driving motivation for the user?
Is it a Cognitive, Piggyback, or Stealth intervention?
Is it a Downstream intervention or a Upstream Intervention?
Is it a Single Agency play or Multi-Agency requiring strategic partnerships?
Is there a strategy for rapid diffusion of enabling technology?
Banny Banerjee, Stanford University
QUESTIONS
Is the segmentation based on motivation, behaviors, and identity?
Is it a Pull or Push intervention?
Will it self propel or do you have to scale distribution?
What is the driving motivation for the user?
Is it a Cognitive, Piggyback, or Stealth intervention?
Is it a Downstream intervention or a Upstream Intervention?
Is it a Single Agency play or Multi-Agency requiring strategic partnerships?
Is there a strategy for rapid diffusion of enabling technology?
Does it piggyback on infrastructure that has already scaled?
Banny Banerjee, Stanford University
QUESTIONS
Is the segmentation based on motivation, behaviors, and identity?
Is it a Pull or Push intervention?
Will it self propel or do you have to scale distribution?
What is the driving motivation for the user?
Is it a Cognitive, Piggyback, or Stealth intervention?
Is it a Downstream intervention or a Upstream Intervention?
Is it a Single Agency play or Multi-Agency requiring strategic partnerships?
Is there a strategy for rapid diffusion of enabling technology?
Does it piggyback on infrastructure that has already scaled?
Does this lead to a new culture and choice architecture?
Banny Banerjee, Stanford University
Banny Banerjee, Stanford University
Potency of Intervention
Banny Banerjee, Stanford University
Potency of Intervention
*
Scale of Implementation
Banny Banerjee, Stanford University
Potency of Intervention
*
Time to*Impact
Scale of Implementation
Banny Banerjee, Stanford University
Potency of Intervention
*
Time to*Impact
Degree of*Integration
Scale of Implementation
Banny Banerjee, Stanford University
Thank you