Non-Traditional Data (NTD): “data that is digitally captured (e.g. mobile phone records), mediated (e.g. social media), or observed (e.g. satellite imagery), using new instrumentation mechanisms, often privately held.”

Digitalization and the resulting datafication have introduced a new category of data that, when re-used responsibly, can complement traditional data in addressing public interest questions—from public health to environmental conservation. Unlocking these often privately held datasets through data collaboratives is a key focus of what we have called The Third Wave of Open Data. 

To help bridge this gap, we have curated below recent examples of the use of NTD for research and decision-making that were published the past few months. They are organized into five categories:

• Health and Well-being;
• Humanitarian Aid;
• Environment and Climate;
• Urban Systems and Mobility, and 
• Economic and Labor Dynamics.

HEALTH AND WELL-BEING

Predicting Health Outcomes with Internet Search Data

Thompson, Matthew, Calvin Chan, Elisabeth Daniels et al. “Systematic review of health research using internet search data.” PREPRINT (Version 1) available at Research Square. June 4, 2024. https://tinyurl.com/2hm86vya. Accessed through https://tinyurl.com/2swfs95z 

• Focus: Systematic review of studies using Internet search patterns to predict and diagnose health conditions.

• Use of Non-Traditional Data: Search engine data (keywords, frequency, timing), both aggregated, anonymized and consented, individual, to identify early symptoms of mental health disorders, cancers, and more.

• Impact: Demonstrated potential for early diagnosis but highlighted challenges like prediction accuracy and privacy concerns.
  
  

Improving Healthcare Access in Lithuania

Vincent, Margaux, and Antti Moisio. “Facilitating Inter- Municipal Cooperation in Lithuania to Provide Better Access to Healthcare and Long-Term Care Services to Citizens.” Development Data Partnership. December 2, 2024. https://tinyurl.com/ycy6sh9r 

• Focus: Addressing healthcare and long-term care (LTC) access challenges in Lithuania’s Tauragė+ functional zone using geospatial data.

• Role of Non-Traditional Data: The OECD's RDG Division used the Mapbox Isochrone API to calculate travel time isochrones for driving and walking to healthcare facilities. These were overlaid with population data to identify accessibility gaps in healthcare and long-term care services.

• Impact: The research team provided roadmaps for improving inter-municipal cooperation, including mobile healthcare services and transport solutions. 

Mobile Technology and Digital Biomarkers for Depression

Zhang, Yuezhou, Amos Folarin, and Richard Dobson. “Exploring Digital Biomarkers for Depression Using Mobile Technology.” Department of Biostatistics & Health Informatics, King’s College London. June 10, 2024. https://tinyurl.com/55us2u6c  

• Focus: Using smartphone and wearable data to track depression.

• Role of Non-Traditional Data: Captured indicators like sleep quality, physical activity, sociability, walking cadence, and circadian rhythm to monitor mental health trends.

• Impact: Showed mobile technologies’ potential impact on depression monitoring and informing personalized interventions.
  
  

Tracking Pandemics with Social Media Data

Shi, Boyang, Weixiang Huang, Yuanyuan Dang, & Wenhui Zhou. "Leveraging social media data for pandemic detection and prediction." Humanities and Social Sciences Communications Volume 11, Article 1075. August 23, 2024. https://tinyurl.com/2s3hjjne.

• Focus: Social media posts as tools for pandemic detection.

• Role of Non-Traditional Data: User-generated content tracked outbreak severity in real time.

• Impact: Demonstrated social media's utility for improving outbreak detection and resource allocation.

Digital Footprints for Population Health

Burgess, Romana, Elizabeth Dolan, Neo Poon, Victoria Jenneson, Francesca Pontin, Torty Sivill, Michelle Morris, Anya Skatova. “Harnessing Digital Footprint Data for Population Health: Collaboration, Challenges, and Opportunities in the UK.” BMJ Health & Care Informatics 2024;31:e101119. September 28, 2024. https://tinyurl.com/5ccuwd5k  

• Focus: Linking NTD with traditional data to predict issues like ovarian cancer and identify public health disparities in the UK. 

• Role of Non-Traditional Data: Linked digital activity (e.g., loyalty card transactions, medication purchases) to health and education records.

• Impact: Highlighted NTD’s role in informing targeted health policies and reducing inequalities in access to health services.
  
  

Meta’s Data for Good Furthers Research into Pathogenic Spread

Ascione, Claudio, and Eugenio Valdano. "How floods may affect the spatial spread of respiratory pathogens: The case of Emilia-Romagna, Italy, in May 2023." MedRxiv. September 23, 2024. https://doi.org/10.1101/2024.09.20.24314056

• Focus: Analyzing how the 2023 Emilia-Romagna floods affected the spread of respiratory pathogens like COVID-19.

• Role of Non-Traditional Data: Mobility data from Meta identified shifts in contact patterns. Researchers used Meta’s colocation maps, which measure the rate at which individuals from different geographic areas are in close proximity, built from aggregated and anonymized mobile device location data with weekly resolution. Geospatial data from Copernicus Emergency Management mapped flood-affected areas, and baseline social contact rates from the SOCRATES project provided context for the observed changes.

• Impact: Demonstrated how mobility data can inform public health planning to mitigate flood-driven epidemics.
  
  

Social Capital and Sexual and Reproductive Health in Africa

Koebe, Till, Theophilus Aidoo, Ridhi Kashyap, Douglas R. Leasure, Valentina Rotondi, and Ingmar Weber. “Social Capital Mediates Knowledge Gaps in Informing Sexual and Reproductive Health Behaviours Across Africa.” Social Science & Medicine 357: 117159. 2024. https://tinyurl.com/ysm972sm.

• Focus: Exploring the effects of social capital on sexual and reproductive health knowledge and behaviors across 33 African countries.

• Role of Non-Traditional Data: Combined Facebook’s Social Connectedness Index, a proxy for social capital, with Afrobarometer and Demographic and Health Survey data.

• Impact: Revealed regional disparities and offers insights into network effects on health outcomes, informing targeted public health campaigns..

Increasing Data Sharing for Social Good: Lessons from Africa’s COVID-19 Response

Amutorine, Morine, Neil Lawrence, and Jessica Montgomery. "Increasing data sharing and use for social good: Lessons from Africa’s data-sharing practices during the COVID-19 response."  Data & Policy, Volume 6, e53. November 18, 2024. https://doi.org/10.1017/dap.2024.43 

• Focus: Examining data sharing practices during Africa’s COVID-19 response by analyzing 74 data-driven initiatives.

• Role of Non-Traditional Data: Mobile phone records and geospatial data were used to track mobility patterns, predict disease spread, and support policy decisions, with privacy protocols and collaborative frameworks ensuring responsible use.

• Impact: Highlighted the importance of robust governance, data literacy, and partnerships to enable sustained data sharing practices that can address broader societal challenges beyond public health crises.

HUMANITARIAN AID

OCHA and Meta: Population Data for Humanitarian Aid

Katch, Kate. “OCHA and Meta: Population Data for Humanitarian Aid.” United Nations OCHA. October 29, 2024. https://tinyurl.com/bdfyuyyw 

• Focus: Partnership between Meta and OCHA’s Humanitarian Data Exchange (HDX) to improve humanitarian planning.

• Role of Non-Traditional Data: AI-powered population density maps provided granular demographic data, enabling targeted responses.

• Impact: The World Health Organization used the maps to identify unvaccinated children in remote Democratic Republic of Congo villages, halting a polio outbreak, while World Vision pinpointed underserved areas for clean water projects, aiding over 1 million people in Rwanda.

Integrating Traditional and Social Media Data to Predict EU Migrant Stocks

Yildiz, Dilek, Arkadiusz Wiśniowski, Guy J. Abel, Ingmar Weber, Emilio Zagheni, Cloé Gendronneau, & Stijn Hoorens. “Integrating Traditional and Social Media Data to Predict Bilateral Migrant Stocks in the European Union.” International Migration Review. December 2024. https://tinyurl.com/446hejdf 

• Focus: Estimating bilateral EU migrant stocks by combining traditional data sources like censuses and Labour Force Surveys with Facebook’s Advertising Platform data.

• Role of Non-Traditional Data: Facebook audience size data, collected from self-reported information used to enable advertisers to target campaigns, provided near real-time, granular insights into migrant population.

• Impact: The statistical model created by the research team generated estimates that closely match available observed data, revealing migration patterns that can inform EU policymakers on mobility trends and migration dynamics and providing further evidence of the utility of incorporating NTD for migration analysis.
  
  

Enhancing Haiti’s Disaster Preparedness with Mobility Data

Flowminder. “Opening up Haiti Mobility Data Platform to Enhance Preparedness and Crisis Response for the 2024 Hurricane Season.” Flowminder News. July 19, 2024. https://tinyurl.com/bdefuf8w. 

• Focus: Supporting disaster preparedness in Haiti by providing updated mobility and population data for humanitarian and development actors.

• Role of Non-Traditional Data: Uses pseudonymized mobile phone Call Detail Records, enhanced with survey data, to track population distribution and mobility trends.

• Impact: Provides insights into population shifts which can aid in assessing exposure to hazards, planning humanitarian interventions, and generating timely crisis reports, contributing to better disaster preparedness and response strategies.

ENVIRONMENT AND CLIMATE

Measuring the Climate Footprint of Tourism in the Nordics

Johansen, Kathrine Lindeskov. “Measuring the Climate Footprint of Tourism in the Nordics with SF-MST.” Centre for Regional and Tourism Research (CRT), Denmark. Presented at 8th International Conference on Big Data & Data Science for Official Statistics, Bilbao. June 2024. https://tinyurl.com/4as5n6yt 

• Focus: Quantifying greenhouse gas emissions from tourism in the Nordics using the Statistical Framework for Measuring the Sustainability of Tourism (SF-MST).

• Role of Non-Traditional Data: Combines tourism satellite accounts (TSA), environmental accounts (SEEA), and big data. For international transport, data sources include Call Detail Records, airport statistics, and Google Maps to estimate distances and emissions by transport mode.

• Impact: Highlights discrepancies in emissions calculations between Nordic countries and emphasizes the need for harmonized methodologies to improve comparability and address data gaps.
  
  

How Energy Data Fuels AI Startups and the Green Transition

Osimo, David, and Anna Pizzamiglio. “How Energy Data Fuels AI Startups and the Green Transition.” The Lisbon Council. 2024 .https://tinyurl.com/55w7f2zm 

• Focus: Exploring how granular energy data supports AI innovation and cleantech growth via the EDDIE consortium’s decentralized, open-source energy data space.

• Role of Non-Traditional Data: Smart meter and IoT data can help optimize energy use, integrate renewables, and stabilize the energy grid. 

• Impact: Examples like Denmark’s Ento and the EU’s Flexidao demonstrate how startups can potentially accelerate Europe’s green transition and foster a competitive cleantech ecosystem.
  
  

Tracking Private Sector Climate Adaptation Finance

Connolly, Jake, Morgan Richmond, William Wallock, Sasha Abraham, Neil Chin, and Chris Grant. “Tracking and Mobilizing Private Sector Climate Adaptation Finance.” Climate Policy Initiative. September 25, 2024. https://tinyurl.com/zhffkjtx.   

• Focus: Developing a taxonomy and machine learning framework to track private sector investments in climate adaptation, identifying and classifying 165 types of climate adaptation activities.

• Role of Non-Traditional Data: Corporate policies,  insurance premiums, and household spending, combined with machine learning tools which capture flows from private actors, including banks, asset managers, and insurers, which are often missed using traditional methods.

• Impact: Improved visibility into private sector adaptation efforts dispels the narrative of limited business models in adaptation. Findings highlight the private sector’s role in bridging the $212 billion annual adaptation finance gap for developing economies by 2030.

URBAN SYSTEMS AND MOBILITY

New Data Sources and Their Impact on Active Traffic Management

National Academies of Sciences, Engineering, and Medicine. "Developing a Planning and Evaluation Guide for Active Traffic Management Strategies. Appendix B: White Paper on New Data Sources and Their Impact on ATM." The National Academies Press.  2024. https://tinyurl.com/nz6wm25u. 

• Focus: Examining emerging data sources for traffic optimization. 

• Role of Non-Traditional Data: Integrating data from connected travelers, vehicles, and advanced tools like sensors and high-definition maps provide real-time traffic insights, while big data tools enable scalable storage, processing, and improved decision-making.

• Impact: Despite challenges, these tools promise expanded geographic coverage and reduced costs for smaller agencies, as well as improved safety and efficiency.
  
  

Urban Freight Data Innovations: ANPR, Delivery Apps, and Truck Pricing

Dablanc, Laetitia and François Adoue. "New ways of collecting urban freight traffic data and applications for urban freight policies and research." Case Studies on Transport Policy, Volume 19. March 2025. https://doi.org/10.1016/j.cstp.2024.101315 

• Focus: Examining new urban freight data collection methods in Rotterdam (ANPR cameras), Barcelona (delivery app), and Brussels (truck GPS data).

• Non-Traditional Data: Automated number plate recognition (ANPR) cameras, delivery apps, and GPS provided real-time freight insights, improving enforcement and parking management, but face hurdles like privacy, data-sharing limits, and inconsistent quality.

• Impact: Better data can inform policy development and implementation in low-emission zones, delivery zone management, and freight monitoring, aiding cities in addressing inefficiencies and sustainability goals.
  
  

Using Facebook Data to Analyze Counter-Urbanization Trends

Duan, Qianwen, Jessica Steele, Zhifeng Cheng, et al. “Identifying Counter- Urbanisation Using Facebook’s User Count Data.” Habitat International 150 (2024): 103113. June 4, 2024. DOI: 10.1016/j.habitatint.2024.103113.

• Focus: Examining counter-urbanization trends in Belgium and Thailand by analyzing population redistribution from urban to rural areas during and after the COVID-19 pandemic.

• Role of Non-Traditional Data: Facebook user count data provided real-time insights into population shifts, supplemented by imputation for missing data.

• Impact: Findings revealed sustained rural migration in Belgium, while Thailand’s rural shifts were temporary. The study emphasizes social media data’s utility in understanding nuanced population trends in different contexts.

Human Mobility Amid Natural Hazards

He, Zeyu, Yujie Hu, Leo L. Duan, & George Michailidis. “Returners and Explorers Dichotomy in the Face of Natural Hazards.” Scientific Reports 14, Article 13184 (2024). June 8, 2024. https://tinyurl.com/mr3ukzn8 

• Focus: Analyzed the persistence of mobility patterns (returners and explorers) in Lee County, Florida during Hurricane Ian.

• Role of Non-Traditional Data: Researchers used anonymized high-resolution mobile phone data, which provided granular, near-real-time insights into human movement patterns.

• Impact: Demonstrated how evacuation behaviors and mobility traits adapt during hurricanes, offering insights for emergency planning, evacuation strategies, and resource allocation for disaster resilience.
  
  

Unlocking Mobility Insights and Open Data

Industry Data for Society Partnership (IDSP). "Year in Review 2024." IDSP. December 2024. https://www.industrydataforsociety.com/uploads/2024/12/IDSP-Year-In-Review-2024.pdf 

• Focus: Telefónica Tech collaborates with National Statistical Offices around the world, using anonymized mobile phone data to monitor tourism, commuting, and urban movement. 

• Role of Non-Traditional Data: Leverages aggregated mobile data to deliver real-time insights into population mobility, complementing traditional statistics.

• Impact: Telefónica’s collaborations demonstrate the potential of anonymized mobile phone data in enhancing our thank you!understanding of population mobility. At the same time, a key outcome is the establishment of an open data ecosystem, where mobility insights are made publicly accessible, fostering innovation and broader societal benefits.  
  

ECONOMIC AND LABOR DYNAMICS

Real-Time Insights into ICT Sector Growth 

Lesher, Molly. 5th International Seminar on Big Data for Official Statistics, 29-31. Xiamen, China. May 2024. https://tinyurl.com/4th8cn7f 

• Focus: Developed a machine learning model to nowcast real-time ICT sector growth.

• Role of Non-Traditional Data: Used Google Trends’ monthly Search Volume Index to complement traditional OECD estimates.

• Impact: Model captured more timely, sector-specific signals that allow for improved measurement precision and relevance.

Quantifying the Effect of Striking on Grocery Store Foot Traffic

Post, Phillip. “Quantifying the Effect of Striking with Picketing on Grocery Store Foot Traffic.” EPJ Data Science, vol. 13, article 59. September 27, 2024. https://doi.org/10.1140/epjds/s13688-024-00495-w.

• Focus: Investigating the impact of striking and picketing on consumer behavior using foot traffic data from 78 striking King Soopers grocery stores in Colorado.

• Role of Non-Traditional Data: Utilized mobile geolocation data from SafeGraph, paired with SARIMA modeling, to quantify foot traffic changes during the 2022 strike.

• Impact: Demonstrated a significant average 47% decrease in foot traffic at striking stores, with a 14% decrease at non-striking locations, providing quantifiable insights into the economic effects of picketing.
  
  

Reflections

The above use cases illustrate that non-traditional data is increasingly being recognized as a valuable resource for addressing complex societal challenges, offering insights that traditional data sources alone often cannot provide. In addition, our curation indicates that:

• Most use cases which we managed to identify originate from academic research or public institutions such as official statistics agencies;

• Health is still the leading sector for re-using non-traditional data such as mobile geolocation, or digital biomarkers;

• Effective partnerships, such as those between Meta and OCHA referenced above; Cuebiq's Data for Good partnerships; the Industry Data for Society Partnership; and the Worldbank’s Development Data Partnership, demonstrate the value of collaboration in unlocking data’s potential.

• Linking non-traditional datasets with traditional ones (e.g., health or census data) amplifies their value, as seen in studies of public health disparities and environmental monitoring.

• Combining non-traditional datasets, such as linking loyalty card transactions with on-line health records in the UK or integrating mobility and flood data in Italy, provides for  more actionable findings.

• Many promising applications, such as using search patterns for early disease detection or leveraging digital footprint data, remain in experimental or pilot stages.

• To move beyond pilots, investments in infrastructure, capacity-building, and long-term partnerships are needed. In particular, data stewards can play a crucial role as the human infrastructure that can support data analysts and researchers in recognizing opportunities for collaboration and structuring access to data for re-use responsibly and systematically.

Back to the Blog