Global Pulse is an innovation initiative of the UN Secretary-General, harnessing today's new world of digital data and real-time analytics to gain a better understanding of changes in human well-being. Global Pulse hopes to contribute a future in which access to better information sooner makes it possible to keep international development on track, protect the world's most vulnerable populations, and strengthen resilience to global shocks.
Global Pulse functions as an innovation laboratory, bringing together expertise from UN agencies, governments, academia, and the private sector to research, develop, test and share tools and approaches for harnessing real-time data for more effective and efficient policy action.
2. contents introDuction
2 closing tHe information gap witH Digital Data “ today’s new world of data holds the promise of giving decision-makers real-time
awareness of how crises are impacting vulnerable populations, and real-time
feedback on whether our policies and programmes are working to protect them.
4 tHe potential of real-time Data together, we will create a more agile, 21st-century approach to global development. ”
-UN Secretary-General Ban Ki-moon
6 from Data to action Decision makers working to protect vulnerable communities are confronting a new challenge: in today’s increasingly
connected world, socioeconomic crises emerge and spread around the globe almost with the speed of natural disas-
ters. At a time when our need for policy agility has never been greater, we have found that our 20th-century tools for
tracking development cannot keep up.
8 a feeDBack loop approacH Global Pulse, an innovation initiative in the Office of the Secretary-General, represents a new opportunity to help
protect the vulnerable through innovation. Global Pulse is partnering with interested governments, UN agencies, aca-
demia and the private sector to explore how emerging technologies and new sources of data can allow Policymakers
to stay ahead of fast-moving crises.
9 leVeraging un system expertise Global Pulse functions as an innovation laboratory for the UN and interested Member States to learn how to harness
new data for development. To that end, Global Pulse is primarily focused on three interdependent areas of work:
Data Research: Exploring innovative methods of combining new types of real-time data with traditional development
indicators to detect early impacts of global shocks.
10 finDing Digital signals
Developing Collaborative Technologies: Assembling a technology platform to house cutting edge tools for sharing
ideas and hypotheses and for analyzing data.
Network of Pulse Labs: Launching a catalytic network of in-country innovation centers that will explore real-time
12 HuncHworks impact monitoring and decision-making at the country-level and will contribute to developing the best practices for
using real-time data in development work.
You can read more about our work in the following pages and on our website: www.unglobalpulse.org
14 pulse laBs We hope you will be inspired to join our effort to harness the tremendous potential of real-time data for the future of
development.
3. iNNovaTioN
closing tHe
analyzing this data for a real-time under- in behavior increasingly leave behind research, develop, test and share tools
standing of their customers. The time trails in data that provide a powerful and approaches for harnessing real-time
has come for us to learn how to use this new tool to help flag early signals of data for more effective and efficient
new data to understand when households stress or incipient harm. policy action. Global Pulse is working
are struggling to make ends meet. to discover real-time proxy indicators of
information gap
To protect communities against the changes in human well being, develop-
When combined with the wealth of tra- impacts of fast-moving crises and keep ing free and open source technologies
ditional mechanisms for data collection global development on track, we need for analyzing data and sharing hy-
that are already in place, this new data real-time information on how com- potheses, and working with interested
holds enormous potential to improve munities are coping with the impacts Governments to establish a network of
witH Digital Data
our understanding of how populations of crises and real-time feedback on Pulse Labs in developing countries.
are impacted by global shocks. Popula- how well programs and policies are
tions under stress change their behav- working. If we are to adapt to this new Today’s ocean of digital data represents
ior. As they cope with lost income or landscape, we need a 21st century an unprecedented opportunity to im-
higher food prices, they change what approach to tracking development. prove the accuracy and agility of policy
they buy and sell. They change how Developing this new approach is the action to protect vulnerable populations
they communicate. They seek different mission of Global Pulse. and strengthen our resilience in a com-
Policymakers in today’s hyper-connected world are grappling with a volatile kinds of information, and they begin us- plex and volatile world. The technology
ing services in different ways. They may Global Pulse functions as an innovation tools to make sense of this data are
environment in which crises in one part of the world emerge suddenly, reverberate look for additional employment, buy laboratory, bringing together exper- available today. The time has come for
around the globe, and amplify the effects of crises already underway. less expensive foods or ask relatives for tise from Governments, UN agencies, us to learn, together, how to take the
financial help. In 2011, these changes academia, and the private sector to global pulse. n
W
e are witnessing an un- earlier—before the onset of the current Waves of compound global shocks
precedented phenomenon: economic downturn. cause households to lose ground in the information gap
socioeconomic crises that their struggle to escape poverty and
move almost with the speed Knowing in close to real time the drive many of those who had escaped Household-level data is so hard to collect that the information being used to track development progress is frequently
of natural disasters. At a time when our “how, where and when” of a crisis back into destitution. Yet today’s tools out of date. We know what happened in the past but not what’s happening in the present. Real-time data could
need for policy agility has never been is crucial for the design of effective do not allow us to track this descent strengthen progress indicators by closing the information gap.
greater, we have found that our 20th- policy responses. While this may be into poverty as it actually happens,
century tools for tracking development obvious in the aftermath of a natural or to catch it early enough to make INFORMATION GAP
cannot keep up. Too often, by the time agile adjustments to our policies and Age of Most Recent Data
we have hard evidence of what is hap- programmes where required. This lack
pening at the household level, the harm knowing in close to real- of timely and actionable information
1 YEAR 2 YEARS 3 YEARS 4 YEARS
has already been done. creates a risk of reversing years of de-
time the how, where and
HUNGER
AND
velopment gains in a matter of months. POVERTY
The recent waves of global shocks— when of a crisis is crucial
food, fuel, and financial—have re- Yet we also face an extraordinary op-
vealed a wide gap between the onset
for the design of effective portunity today. The mobile digital
EDUCATION
of a global crisis and the availability policy responses. revolution of the past decade now has GENDER
of actionable information for decision us swimming in an ocean of real-time EQUALITY
makers. Traditional statistics have been disaster, it is also increasingly evident data, most of which did not exist even
effective in tracking medium- to longer- in the context of global events such as a few years ago. From the content of CHILD
MORTALITY
term development trends, but—given the recent financial crisis. Finding out social media sites to the data we leave
the latency of the data generated—are today how a community began coping behind in communicating, transacting
of little use to decision makers trying to with a crisis two years ago is generally business, and searching for informa- MATERNAL
HEALTH
help vulnerable populations cope with too late to prevent longer-term dam- tion, people everywhere are generating
fast moving crises. Much of the data age. It is clear that the price a country vast quantities of data, for free, merely
HIV AND
used to track progress toward the Mil- pays for not knowing how specific seg- by going about their daily lives. MALARIA
lennium Development Goals (MDGs), ments of its population are coping with Source: Millennium Development Goals Report 2011
for example, dates back to 2008 or shocks is high. Private sector companies are increasingly
2 gloBal pulse Harnessing innovation to protect the vulnerable 3
4. D
ata from blogs and other online ing, whether they are using services like and tackle questions like these.
discussions together with “data mobile banking or whether users are
exhaust,” (the digital trails we moving en masse from areas affected Additionally, the Global Pulse technol-
leave behind as a by-product of by crises like drought, disease or ogy team is designing a tool called
daily activities) are increasingly provid- shortages of food or services. HunchWorks that will allow development
ing the kind of information that previ- experts to share hypotheses about wor-
ously could only be gleaned from face- Aggregated e-commerce data that can rying trends in data, link to additional
to-face surveys. An overview of the new show changes in buying and selling evidence and debate and verify those
data landscape reveals that real-time behavior e.g. people cutting back on “hunches” within a social network of
analysis of such data will be integral to non-essential purchases or buying more trusted colleagues. Those colleagues may
policy planning in the future. medicines than in previous months. bring in additional evidence to support or
to refute the hypothesis. (For more, see
Recognizing this potential, the Executive Feeds from social media sites where HunchWorks on page 14).
Office of the Secretary-General launched people voluntarily share information
the Global Pulse initiative in 2009 to about their health and finances. In “The projects we are undertaking here
explore how this new data could help the aggregate this can tell a lot about the at Global Pulse are designed to bring
United Nations’ development community well-being of communities. global experts in the emerging field of
obtain a more real-time understanding data science together with those who
of populations’ emerging vulnerabilities. What makes the billions of pieces of are best positioned to understand the
“The United Nations is uniquely placed data culled from these sources use- needs of decision makers in developing
to leverage these new data sources for ful—in essence what transforms this countries,” says Eva Kaplan, Programme
the global common good,” says Zazie new data into insightful information—is Specialist. “Through these collaborative
Schafer, Deputy Director of the initiative. expert human analysis. Defining frame- projects, we are ensuring that the specif-
works for such analysis, and developing ic needs of government decision makers
The term “real-time data” indicates tools when necessary, is at the core of are at the forefront of how we understand
little or no delay between data collec- Global Pulse’s work. the opportunities in new data.”
tion and dissemination. In the context
of Global Pulse, real-time data refers The future use of real-time data in de- To get a better picture of the potential of
to data that is available soon enough to velopment work rests on learning how to real-time data, Global Pulse teamed up
initiate policy actions to prevent harm. effectively and efficiently mine raw data with PSFK, a research consultancy spe-
Some of the new data sources that for actionable information, i.e. informa- cializing in emerging innovation trends,
iNNovaTioN Global Pulse will be looking at include: tion that can be useful in programme to explore some of the best examples of
tHe potential of
planning. To be successful it will be real-time data use. The resulting publi-
Online news and media outlets where necessary to answer some big ques- cation, The Future of Real-Time, high-
trending topics can reveal what areas tions, paramount among them: What are lights key areas of opportunity including
are of concern to readers. the meaningful signals of distress for a instant mapping, mobile communities,
given community? Can those signals be contextual cartography, intelligent infra-
real-time Data
Data generated from information and inferred from the data? Once detected, structure as well as the institutional use
communication technology (ICT) what do they reveal about well-being? of real-time technologies.
devices like mobile phones and “smart And how can they be useful to policy-
phone” platforms, that can show how makers? Global Pulse has launched sev- Look for more on Global Pulse’s re-
populations are changing their spend- eral research projects to further define search projects in this publication. n
The use of real-time data is providing Governments, corporations and organizations
finD out more online:
with rapid ways of detecting and responding to the needs of the populations they For a blog post on The New Data Landscape that explores new sources of real-time data, visit www.unglobalpulse.org/blog/new-data-
serve without requiring significant new infrastructure. landscape. Find the PSFK Future of Real-Time report at www.unglobalpulse.org/projects/future-real-time-report
4 gloBal pulse Harnessing innovation to protect the vulnerable 5
5. iNNovaTioN
from
SOCIAL MEDIA
SOCIAL MEDIA
PRIVACY
PRIVACY ANALYSIS
ANALYSIS ACTIONABLE DATA
ACTIONABLE DATA
NEWS FEEDS
NEWS FEEDS
Data
A A A A
MOBILE PHONE DATA
MOBILE PHONE DATA
G G N N
G G O O D D V V
R R N N E E E E
E E Y Y T T R R
to
GEOSPATIAL DATA
GEOSPATIAL DATA
G G M M E E I I
A A I I C C F F
T T Z Z T T Y Y
UN DATA
UN DATA
E E E E
action
GOVERNMENT DATA
GOVERNMENT DATA
CITIZEN REPORTS
CITIZEN REPORTS
1. Data 2. priVacy 3. analysis 4. action
Big Data can yield information that Real-time data originates both Once aggregated, data is fully Detection of anomalies in the data Real-time data enables decision
from new technologies as well as anonymized to protect privacy and trigger follow-up investigation makers to protect vulnerable
speeds up decision-making while still from new approaches to mining to facilitate analysis that is focused and verification using both new populations with more agile and
protecting individual privacy. traditional data. on the community, not on the technologies as well as traditional targeted responses.
individual. analytics.
G
lobal Pulse is built on the step 1. Data collection policymakers and researchers. community level indicators; the identity step 3. analysis sion makers to target and refine their
understanding that today digital The Internet supports the flow of of individuals plays no role whatsoever. Of course, it is not enough just to have interventions more efficiently than is
data is being generated at an massive amounts of user-generated Data is also generated from govern- Global Pulse does not handle data with this data. Most important is developing possible today. In a fast-moving, hyper-
unprecedented rate and can information such as news, social media, ments and agencies tracking the personally identifiable information, and an understanding of how it can be use- connected, and volatile world, public
hold real insights about changes in blogs and e-commerce sources among progress and effectiveness of their pro- its country-level research is conducted ful. Part of the equation is exploring policy must be agile and continuously
population well-being. Knowing how many others. grammes in the form of surveys, hotline within the context of national data how new data can be integrated with adaptive if it is to protect vulnerable
vulnerable populations are changing records, and inquiries. When people’s privacy laws. Because Global Pulse and the traditional sources of data. populations from shocks, keep global
their behavior, especially during the Populations also generate and share needs change, they change how they its research partners are interested in development on track, and meet the
unfolding of a global crisis, can help data as by-products of their daily lives. participate in such programmes, and the collective picture that emerges from Another important factor is developing changing needs of citizens.
decision makers accelerate and target For example, as communities in the those changes are captured in the data. data sources, private and public sector local expertise, because recognizing the
programme planning and policy mak- developing world increasingly adopt and Global Pulse calls this data “pro- partners must anonymize parts of the signs of stress in a particular population Global Pulse is developing methodolo-
ing in ways that can have great im- use new technologies—particularly mo- gramme exhaust.” data that are not statistically important often depends on familiarity with the gies with which decision makers will
pact, especially in the fight to protect bile phones and services provided over to the research. local context. In addition, the availability be able to integrate these data streams
development gains. mobile—they generate ambient data step 2. ensuring priVacy of data and the types of signals the data into their existing monitoring efforts
or “data exhaust.” Much of this data is anD anonymity In fact, the information Global Pulse is yields will differ from country to country. to detect anomalies, form hypotheses,
Global Pulse’s work is focused on mak- not publicly available but is owned— Mining data for the pulse of commu- exploring is most valuable in aggregate, gather evidence, verify the underly-
ing data actionable for decision makers and held closely—by the private sector. nities does not equate to monitoring when it can show how a population in a step 4. action! ing causes and take action to improve
to use in shaping programmes and Global Pulse is developing innovative individuals. Global Pulse takes privacy particular region is collectively chang- At Global Pulse, we recognize that the well-being of the populations they
policies in real-time. Moving from data public-private partnerships to allow very seriously. Global Pulse’s data ing its behavior in response to global analysis of patterns within new streams serve, especially in addressing the
to action requires several steps: analysis of this data to be shared with research is focused solely on analysis of shock or a policy change. of real-time data can enable deci- impact of global shocks and crises. n
6 gloBal pulse Harnessing innovation to protect the vulnerable 7
6. iNNovaTioN ReseaRCh
leVeraging un
high frequency, household-level data
available through surveys and quick participating agencies
BASELINE
impact assessments.
system expertise
The UK’s Department of
A model for this process has been used International Development (DfID)
for years in the field of public health to and the Government of Sweden
a
G
monitor for outbreaks of infectious dis- provided funding for the RIVAF
eases. When public health officials see lobal Pulse is dedicated to formulation and advocacy. projects. The following eight
feeDBack
people not coming in to work or school, developing new information agency research projects were
ACTION DETECTION buying certain drugs, and showing up streams to protect vulnerable New methodologies and statistical chosen for exploration:
at clinics with fever and headache, they populations in the context of models were tested for their general
loop
may suspect an outbreak of flu. Detec- increased global shocks. This requires applicability. unDp/unicef: Monitoring
tion of these anomalies triggers a process strengthening the ability to understand Households’ Coping Strategies
of rapid investigation (collection of speci- the signals in new data; it also requires The project participants met at a con- During Complex Crises
approacH
mens) and verification (laboratory testing) understanding the dynamic complex ference in New York City in July 2011
that leads to early and effective response. processes that determine how people to discuss the findings of their research unesco: Reviewing the Impact
are impacted by crisis. Thus, one of and to identify common challenges and of the Crisis on Education in
The framework, which draws on similar Global Pulse’s first undertakings, the lessons learned. They also discussed Selected Developing Countries
lines, will be developed and tested Rapid Impact and Vulnerability Analysis how common research challenges
through Pulse Labs in consultation with Fund (RIVAF), was dedicated to lever- could be addressed using new data, unfpa: The Impact of the Global
VERIFICATION INVESTIGATION government, UN agencies, academia aging the expertise already in the UN. and—building on the findings of the Financial Crisis on Reproductive
and the private sector and will involve: Global Pulse invited UN agencies to first round of RIVAF projects—explored and Maternal Health in Colombia,
submit project proposals designed to ideas for a future round of research. Ethiopia and Jordan
Proposing and investigating hypotheses examine the multiple ways that those
At Global Pulse we say where there is a digital smoke about the causal factors driving these impacts manifested. Participants were In particular, the collaborations unicef/wfp: Food and Nutrition
collective behavior changes based on encouraged to submit plans that could confirmed the existence of significant Security Monitoring and Analysis
signal—an unexplained change in how people analysis of additional contextual be implemented in partnership with data gaps, affirming the need for Systems: A Review of Five Coun-
communicate or use digital services—there may also be information; at least one other agency. In the end access to compelling real-time data, tries (Indonesia, Madagascar,
11 UN agencies undertook eight rapid which is one of the core goals of the Malawi, Nepal and Zambia)
fire: a need for action. Sharing of hypotheses within trusted impact assessments that covered 38 Global Pulse initiative. Participants
networks where experts share informa- unoDc: Monitoring the Impact of
U
countries and four cities. expressed strong interest in further
nderstanding quickly where tive framework for combining traditional tion to support or refute hypotheses; collaborating with Global Pulse to test Economic Crisis on Crime
the fire is requires additional indicator-based approaches to track- The collaborations added value to how new digital data sources could
reporting and analysis. So does ing development with new kinds of Testing promising hypotheses by agency work in multiple ways: help fill existing knowledge gaps. unoDc: Smugglers and Vulner-
understanding what the digital real-time data to pick up anomalous soliciting feedback—perhaps via SMS Everyone agreed that a future round able Migrants in Central America
smoke signal means, and what action changes in behavior. text messages—from trusted contacts Several projects developed online tools of RIVAF funding would be ideal as and Mexico
is required. Smoke signals aren’t hard within the local community, such as for data transfer, expert collaboration it would facilitate further innovative
evidence. They aren’t actionable infor- This involves analysis of anonymous local government, teachers, community and analysis of large data sets. collaborations between UN agencies, unoosa/uneca/unep: A Visual
mation by themselves. But where they and aggregated digital data to detect health workers, or radio station hosts; as well as with leading academic and Analytics Approach to Under-
appear, something may be happening abnormal patterns in populations’ be- RIVAF research findings provided inputs public and private sector experts in standing Poverty Assessment
that policymakers need to know more havior, and combining it with data that Verifying hypotheses and gathering for agency programming, policy the field of real-time data. n through Disaster Impacts in Latin
about – and quickly. provides context for multi-dimensional information required to design appropri- America and Africa
analysis, cross-validation and verifica- ate responses through targeted, agile
What is needed is an approach that rap- tion. Such contextual data includes: collection of new household-level data; finD out more online: unwto/ilo: Economic Crisis,
idly detects the earliest smoke signals, (a) macro level information (e.g., For more information about RIVAF, including full descriptions of the parameters and Tourism Decline and its Impact
supports verification with contextual in- macro-economic indicators); (b) trend Taking action by working with UN objectives of the projects and their findings, go to our projects page at www.unglobal- on the Poor: An analysis of the
formation and additional evidence, and data for key socioeconomic indicators agencies to support governments in pulse.org/projects. effects of the global economic
informs more timely policy responses. (e.g., MDG data); (c) real-time infor- implementing policy responses and crisis on the employment of poor
mation on specific contextual variables using this new monitoring approach to Download the RIVAF Partners Final Report at www.unglobalpulse.org/research/reports-and- and vulnerable groups in the tour-
Global Pulse is developing an innova- (e.g., weather, price data); and (d) track results. n resources ism sector
8 gloBal pulse Harnessing innovation to protect the vulnerable 9
7. ReseaRCh
finDing Digital “multidisciplinary collaborative projects offer great opportunities for discovering
patterns that will lead to new insights useful for impact monitoring and development”
signals
- Miguel Luengo-Oroz, Data Scientist for Global Pulse
can social media mining add Depth to news awareness & emergent information
unemployment statistics? monitoring
How do you find indicators in big data? How do you know which of those signals are
partner: sas (www.sas.com) partners: complex systems institute of paris (isc-pif), formism,
relevant enough to warrant further investigation? These questions are at the core of ifris, cams, crea, inra-sens.
Global Pulse’s research. How does combining social media data with tradi-
tional data sources amplify understanding of the This multi-partner project analyzes thousands of news
I
experiences of the unemployed? In collaboration with items related to food security issues in French language
n the past year Global Pulse has teamed up with some of methodologies for tracking and monitoring development. SAS, this project analyzes conversations from the media (national and local). It explores how news around
the world’s leading research firms and institutions to en- open social web for early signs of stress among the particular issues gains salience, evolves over time and
gage in an ongoing series of research projects designed to “Multidisciplinary collaborative projects between Global unemployed in order to better understand their cop- answers questions such as: Is an issue (e.g. biofuels and
test the feasibility of real-time data collection and analysis Pulse and private companies and academic researchers are ing strategies for dealing with their unemployment. food security) attracting more or less attention with time?
to find new indicators of population well-being. a fantastic approach for adapting new tools and methods for Another goal is finding out how the unemployed Do discussions around any issue (observed at a given
development. This innovative ecosystem offers great oppor- perceive their status, e.g. whether they are collec- time) stem from how past related issues are framed, or
Results from the initial series of projects have underlined tunities for discovering patterns that will lead to new insights tively optimistic or worried about the future. For this do they emerge distinctly? An online interface allows
several promising approaches to improving monitoring of un- useful for impact monitoring and development,” says Miguel purpose we explore correlations between quantitative users to visualize and map key themes and news entries
employment and food security. Findings from these projects Luengo-Oroz, Data Scientist for Global Pulse. indictors produced with the latest text analysis meth- and shows what issues related to food security are of
will help to establish new methodological approaches for odologies and official unemployment statistics. greatest concern in the media at any given time.
analyzing real-time data, identify tools which can provide a An overview of some of these exciting collaborations follows:
faster and clearer understanding of population behavior dur-
ing periods of stress, and contribute to the development of
new proxy indicators for real-time tracking of development.
global snapshot of well-Being: exploring the comparing global prices to local twitter & perceptions of crisis-related stress
Some of the projects explored ways to produce high frequen- potential of global mobile surveys products: real-time e-pricing of bread
cy data. For example, mobile phones provide a powerful plat- partner: crimson Hexagon (www.crimsonhexagon.com)
form to reach people and to get real-time feedback directly partner: Jana (ww.jana.com) partner: pricestats (www.pricestats.com)
from communities. Other projects explored ways to mine, How can micro-messaging services—particularly,
aggregate and analyze online data—such as food prices or Global Pulse is working with Jana to test the potential of Inflation of the cost of basic food items disproportion- Twitter data—be used to provide analytic insights
news trends—in real time. quickly deployed mobile surveys, over SMS, in multiple ately affects those who are struggling financially. This about how people perceive issues related to food,
countries. The objective is to complement and strength- project, a collaboration with PriceStats researchers, fuel, housing, and the economy? In collaboration
Global Pulse is also exploring the ability to tap an entirely new en our agility in the way we do household surveys, reach- entails the construction of a daily “bread price index” for with Crimson Hexagon, we measured the degree to
source of information: the global conversation, or social media. ing thousands of responders in a few days to provide six Latin American countries to show how a tool might which cost pressures around these core items are
This type of data can help us “listen” to a populations’ own global snapshots. This methodology tests the feasibility be created to identify real-time changes in the prices discussed in online conversations. The research
perceptions of their changing circumstances. The methodolo- of conducting longitudinal surveys over time. of essential food items. The tool measures daily retail analyzes trends in these topics in conjunction with
gies needed to explore this massive unstructured data from so- prices through online supermarkets and combines this keywords such as “afford,” showing how the volume
cial media sources require the use of the latest analytical tech- information with the availability of products to create the and topics of the conversations change over time
niques, including data mining, machine learning algorithms daily price index. This has the potential to help us detect reflecting populations concerns.
and automated text processing. Quantifying the perceptions finD out more online: excessive price volatility on people’s wallets in real-time.
of populations presents a powerful opportunity to understand To follow the progression of these and other projects, visit our
a population’s behavior in a way that complements existing projects page at www.unglobalpulse.org/projects
10 gloBal pulse Harnessing innovation to protect the vulnerable 11
8. TeChNoLogy
HuncHworks
cally analyze evidence attached to
hunches. It will also stimulate knowl-
edge sharing and debate by recognizing
when discrete groups of experts are
working on the same issues and inform-
ing them of each other’s existence.
A research platform that combines observation, social “HunchWorks is like simultaneously
being connected to the smartest, most
networks and big data to identify crisis impacts in real-time. POSSIBLE OUTCOMES informed people you know and the
POSSIBLE OUTCOMES
smartest, most informed people you
IMMEDIATE
don’t know,” says Sara Farmer, Chief
a researcH scenario ACTION
IMMEDIATE Platform Architect for Global Pulse’s
ACTION
A development professional working in a technology team.
Government ministry, UN agency or re-
search organization notices signs of dete- In several ways, HunchWorks will virtu-
FURTHER
riorating social conditions in a vulnerable RESEARCH
FURTHER
ally recreate the working dynamics of
community. As she gathers content from RESEARCH a Pulse Lab, the in-country innovation
household surveys, online news, social centers Global Pulse plans to facilitate
media, and satellite imagery, she grows in the next five years, where diverse
DISMISS
concerned that she may be witness- HUNCH
DISMISS experts work to combine early evidence
HUNCH
ing the start of a harmful longer-term from a variety of sources to support
trend. So she starts to document her more agile and targeted policy action.
observations and gather evidence. She (For more, see Pulse Labs on page 16
also starts to reach out to other credible of this publication)
observers in the region to find out if they
are seeing anything similar. open source DeVelopment
Collaboration is also integral to the
Responses are mixed and her contacts creation of HunchWorks itself as the
are limited to friends within the local 1. oBserVe 2. sHare 3. Verify 4. act tool is being developed as open source
agency and a few local community After a potentially destructive trend is The hunch is shared with trusted networks on Users are able to evolve their collective If there is sufficient consensus that a hunch software with support from a cadre of
figures that she has met. She has seen identified through observation, research or the HunchWorks platform. Users from within understanding through ongoing discussion, a has been confirmed, tools built into HunchWorks volunteer programmers, designers and
this type of pattern in the data before. first-hand accounts, a hunch is posted with these networks provide feedback on the hunch growing body of evidence and analytical tools will allow users to take appropriate action testers. Dedicated volunteers meet
She has a hunch that the trend is a description, location data and any relevant and evidence posted. They are also able to built into HunchWorks. e.g. bundle evidence into reports and activate weekly at Global Pulse’s New York of-
potentially very damaging, but due to evidence. contribute additional evidence. networks for advocacy. fices for “code jams” where they develop
the informal, unverified nature of her and test components and work out kinks
observations and the limited feedback in the system. The code jams have also
of colleagues, she knows that it will be enabled Global Pulse to build a database
difficult to make the case this early to a new tool available early on but was so inconclu- in additional evidence supporting or on a topic, e.g. food security or health of expertise, as many of the volunteers
her superiors for immediate action on The Global Pulse technology team’s sive at the time that no one was willing refuting her conclusion, provided feed- processes, and their respect for each have day jobs at technology companies.
behalf of her agency. answer to this problem is HunchWorks, to sound the alarm publicly. back about her observations, and sug- other’s work. A community of experts
a social network tool for gathering evi- gested other experts to invite into the may have members from different gov- “The open source method allows us to
She knows there must be others observ- dence, analyzing data and peer review Using HunchWorks, the researcher conversation. In effect, HunchWorks is ernment ministries, UN agencies, disci- develop technology that can be useful
ing similar signs who could help gather all at once. could have quietly composed a hunch the world’s first social network for hy- plines or expertise. They may contribute to everyone, gives us access to exper-
more evidence if they knew what to look outlining her concerns. After attach- pothesis formation, evidence collection, empirical evidence from scientific tise that we wouldn’t normally have
for. She also recognizes that there are HunchWorks is based on the recogni- ing relevant evidence, she would have and collective decision-making. research or anecdotal or observational and helps us develop the local tech
experts with deeper subject matter exper- tion that all too often, by the time we shared it informally through the system, evidence from field work. Or they may community,” says Farmer. n
tise who could review her findings. But have hard evidence of what is hap- where it would be visible only to a Researchers who use HunchWorks are help facilitate collaboration, whether it’s
how can she identify and engage these pening, it’s too late to act. Yet after a trusted network of expert colleagues in part of a self-selecting virtual peer com- building systems or moderating debate. finD out more online:
people quickly enough to stay abreast of crisis, the experts inevitably point to relevant communities of practice. Her munity in which their connection is a HunchWorks includes computational See www.unglobalpulse.org/technol-
unfolding events on the ground? bits of fragmentary evidence that was colleagues might in turn have brought deep common interest in and expertise reasoning features that will automati- ogy/hunchworks
12 gloBal pulse Harnessing innovation to protect the vulnerable 13
9. PuLse Labs
A
s the recent global financial man capacity required to harness incubators and innovation
crisis unfolded, it became real-time data for policymaking. spaces—with the twist that
clear to in-country decision Global Pulse is focused on build- they connect to a global network
makers that they did not have ing a small, highly integrated of United Nations’ needs and expertise,”
a real-time understanding of how the network of ten labs in the next five says Christopher Fabian, Global Pulse
crisis was uniquely affecting their own years. So far the Pulse Lab track has Advisor on Innovation (ad interim).
populations. What was needed was the established Pulse Lab New York and
ability to quickly—and as accurately is in the process of developing Pulse It is clear that the most important
as possible—get continuously updated Labs in Uganda and Indonesia. success criterion for Global Pulse is
snapshots of how populations were be-
ing impacted so that policy responses
could be targeted to their most current “pulse labs are the heart of the global pulse
situations. Traditional data collection
methods like door-to-door household
initiative, and we need government support
surveys that can take months or even to bring the vision to full life,”
years to complete are woefully inad-
equate for this task. Finding out today - Makena Walker, Global Pulse Partnerships Adviser
how a community began coping with
a crisis two years ago (or even two
months ago) could mean it is too late to By mining digital data for correlations Member State participation in the Pulse
institute the changes aimed at prevent- with traditional statistical indicators Labs track. “Pulse Labs are the heart of
ing longer-term damage. used to track development, Pulse Lab the Global Pulse initiative, and we need
staff will work to identify patterns that government support to bring the vision
Global Pulse was created from the were being left behind in digital to full life,” says Makena Walker, Global
understanding that what is lacking in data—in real time—as communi- Pulse Partnerships Adviser.
the world of development planning lies ties struggle with the impacts of
in the world of real-time analytics that crises or benefited from govern- pulse laBs in action
is already being employed by the private ment programmes. Once identi- With guidance from Government
sector. Companies use real-time data fied, these patterns will be investigated stakeholders, Pulse Labs will focus
to help them understand the changing thoroughly as potential real-time proxy their research efforts on addressing
needs of their customers and respond by indicators of well-being that could be critical information gaps in under-
designing more agile business platforms. monitored on an ongoing basis to sup- standing the needs and vulnerabilities
However, applying real-time analytics port decision makers. of at-risk populations. After an initial
and digital data to development chal- process of identifying existing data,
lenges is relatively uncharted territory. Pulse Lab New York, which functions as public and private sector sources of
It is not yet clear which of the many and a research and development space and a relevant digital data, and key partner
pulse laBs
varied new data sources will prove most facilitator for partnerships with multi- organizations, Pulse Labs will pursue
valuable for tracking changes in human national private sector data providers, a focused research agenda based on
well-being, nor is it clear which analyti- spearheads Global Pulse’s innovation retrospective analysis. This process
cal methodologies and technology tools in real-time monitoring and leads the consists of analyzing anonymized and
will be required to transform this data design of the technology platform and aggregated digital data generated by
into actionable information. tool kit design. It is currently serving as communities over the past few years
the knowledge hub for Global Pulse’s (e.g., mobile phone calling data,
Pulse Labs are networked, country-level innovation spaces where government experts Accordingly, Global Pulse has elected work, ensuring that as other labs develop money transfer records, online news,
to adopt a joint exploration approach they will be able to access and build on social media content) and comparing it
and UN agencies can experiment with trusted partners to develop methods and tools in which interested Governments each other’s innovations. “Labs provide a with robust statistical evidence of what
for real-time impact monitoring, collaborative analysis and decision-making. partner at the country level with UN physical space for active learning, shar- these communities were experiencing
agencies to establish a network of in- ing and, importantly, creating solutions at the household level during the same
novation centres, or Pulse Labs, that in collaboration with host governments. period (e.g., household survey data,
will develop the approaches, tools, These instruments are modeled on the unemployment figures, food price fluc-
public-private partnerships and hu- success of private sector and academic tuations, disease outbreaks).
14 gloBal pulse Harnessing innovation to protect the vulnerable 15
10. In order to support ongoing research experts from the private sector. Pulse “We’ve learned so much about what it
and contribute to future capacity for Lab teams will coordinate virtually with takes to apply the global vision locally.
real-time monitoring, Pulse Labs will one another as a network for sharing It will certainly make establishing other
How pulse laBs
cultivate strong relationships with a the technological and methodological pulse labs a lot easier.”
variety of local stakeholders through innovations they each produce.
events, campaigns, and other forms of More recently, the scoping phase for
outreach. While still in its early stages, the Pulse Lab Jakarta started in October
work
experience of Pulse Lab Kampala is of 2011. The UNDP country office in
Partnerships with promising private sec- instructive. In December 2010, two Jakarta served as the facilitator. In
tor data providers such as mobile phone innovations officers were posted in partnership with the Government of
carriers and microfinance institutions Kampala to do the initial scoping for Indonesia, the Lab is scheduled to
will be established in ways that allow the lab. UNICEF has an active innova- start operations in June 2012. The
these organizations to share analysis of tions unit in Uganda, and offered to Lab’s objectives are aligned to the
their data without compromising either host the pre-launch phase of Pulse Lab United Nations Partnership Develop-
Pulse Labs operate as hubs where government and UN country teams individual privacy or corporate com- Kampala. The Global Pulse team has ment Framework covering 2011-2015.
can work together with partners such as data analysts, academia, petitiveness. Community leaders will worked to develop partnerships with Two proof-of-concept research proj-
open-source technologists, the private sector and civil society to provide insight into what context-specif- the government and private sector, un- ects—one on mobile data, the other
explore ideas and to design and develop tools that can be used ic changes occur in local communities derstand information gaps and priori- on Twitter—should be completed by
to recognize signals in data exhaust. impacted by crises and suggest ways ties, identify potentially valuable data July 2012.
with guidance from government stakeholders, pulse labs will focus their
research efforts on addressing critical information gaps in understanding the
needs and vulnerabilities of at-risk populations.
GOVERNMENT UNITED
NATIONS
to engage citizens at the grassroots sources, and develop a strategic plan Throughout this process, Global Pulse
level in the process of reporting these for the launch of the Lab. The Lab will collectively put together a handbook
changes when they first appear. will collaborate with key government or “cookbook” for establishing a Pulse
departments such as the National Lab, which will include mappings of
UN agencies and national institu- Planning Authority (NPA) of the Minis- possible partners, staffing needs and op-
PRIVATE SECTOR FOUNDATIONS tions will identify existing programmes try of Finance, Planning and Economic erations objectives. This will be a living
AND CORPORATIONS AND NGOs that could benefit from incorporating Development, the Prime Minister’s document that can be shared widely to
analysis of real-time digital data. Open Office, the Uganda Bureau of Statistics encourage and strengthen similar initia-
source software developers will assem- and other sectoral institutions. The tives in countries and agencies that are
ble, translate, and enhance the technol- private sector, including two mobile not directly involved in Pulse Labs. The
ogy toolkit required for mining digital phone carriers, has expressed interest cookbook will capture best practices and
data and providing actionable data in partnering with the Lab. Other key lessons learned, therefore the knowledge
visualizations for government decision stakeholders include Makerere Univer- base—developed from ongoing inputs
makers. Secondments from Government sity and several social enterprises. from the Pulse Lab network—will be
OPEN SOURCE LOCAL ministries and UN agencies will com- available to other interested countries to
TECHNOLOGISTS ENTREPRENEURS prise a significant fraction of the team, “The Ugandan experience has been jumpstart the development of their own
as will researchers from academia and an eye-opener,” says Makena Walker. crisis impact monitoring capacities. n
COMMUNITY
ACADEMIA
GROUPS
finD out more online:
To learn more about Pulse Labs, go to our website www.unglobalpulse.org/pulselabs. Also check out our multimedia page to watch
videos from Uganda. For information about establishing a Pulse Lab, contact info@unglobalpulse.org
16 gloBal pulse Harnessing innovation to protect the vulnerable 17