GRAAM

Monday, June 15, 2026
1:53 pm
Linkedin X-twitter Instagram Facebook Youtube

Blogs/

Governance Efficiency: Transitioning from Departmental Silos to Integrated Data Linkages

  • Embark India Development Fellowship (EIDF)
  • June 15, 2026

Governance Efficiency: Transitioning from Departmental Silos to Integrated Data Linkages

– Mehakk Bhardwaaj, Embark India Development Fellow 

Introduction

In a remote district, a rural health worker noticed a heartbreaking cycle: children treated for Severe Acute Malnutrition (SAM) at a Nutrition Rehabilitation Centre (NRC) were relapsing just weeks after discharge. Clinically, the “system” worked- the child was stabilised. However, the system was unaware of the fact that for the same family,  the family’s Ration Card had been deactivated due to a technical mismatch in the Food Department’s database. Because of this “silo” problem, administratively, the outcome could not be truly delivered. While India has built a world-class digital spine, our administrative datasets need integration. One department tracks clinical recovery; another tracks food security. Because these systems could not communicate with one another, the child remained vulnerable to recurring malnutrition.

This composite case illustrates a pattern documented across multiple field studies. This article explores the challenges facing welfare delivery and governance systems, and discusses how integrated administrative datasets may help address some of these concerns.

 Context & Background: The Rise of Administrative Datasets

With the rollout of DPI, India created a system where data is generated at every interaction, from a Rs. 10 tea stall payment to a digital vaccine certificate. This has created a “data-rich” environment where even small-scale entrepreneurs can prove their creditworthiness through transaction history. We are now in the era of “Administrative Datasets”, vast pools of real-time information generated during daily operations, such as the 1.3 billion Aadhaar records and the billions of transactions processed through the UPI (Ministry of Finance, 2024).   However, as highlighted in the World Development Report 2021: Data for Better Lives, the value of data is not in its collection, but in its interoperability. In simple terms, interoperability is the “universal translator” of the digital world. It is the ability of different information systems, devices, and applications to access, exchange, and cooperatively use data in a coordinated manner. Interoperability is ensured with dataset harmonisation. Without it, the data is like a library where every book is written in a different, secret code that only its specific author can read.  For example, it is desirable to have interoperability between the Unified District Information System for Education Plus (UDISE+) with a labor database (e-Shram) to see if vocational training is actually leading to formal employment (World Bank, 2021).  For example, Odisha has shifted its vocational success metrics from mere enrolment to long-term “livelihood transitions,” mandating a 6-to-12 month tracking period for all graduates placed via the Placement Linked Training Program (PLTP) (Odisha Skill Development Authority, 2026).

The  Problem 

The primary challenge is not a lack of digital records, but a lack of a unified, automated data-sharing framework. This fragmentation and “Data Silos” can lead to:

  • Citizen Exclusion: Marginalised citizens lose benefits because updates in one system (e.g., a change of address in Aadhaar) do not propagate to others (e.g., the Ration Card).
  • Analytical Waste: Data is  stored in silos with incompatible formats and inconsistent definitions across ministries and departments, making integration manual and time-consuming. A single category of data errors in LPG subsidies required two years of manual reconciliation to resolve, and errors or duplicate entries inflate welfare budgets by an estimated 4–7% annually (NITI Aayog, 2025).
  • Inefficient Targeting: Major welfare schemes like PMAY-G, PM-JAY, and MGNREGA continue to rely on the 2011 SECC for beneficiary identification ,while India awaits its next  census. (Social and Political Research Foundation, n.d.; The Secretariat, 2025).

Current Approach: The “Point Solution” Trap

While the vision for integrated data is clear, the way our current systems are built often prevents this from happening. Most government datasets  are currently managed through “Point Solutions”, standalone digital tools designed to solve one specific task in isolation, rather than addressing the citizen’s broader journey. This fragmented approach creates several hurdles:

  • Manual Verification Loops: In the agriculture sector, while YES-Tech has significantly improved satellite-based yield assessment under PMFBY, and state land records have been integrated with the National Crop Insurance Portal, some states continue to face challenges with land-record integration, digitisation of CCE data, and capacity constraints, meaning delays in claim settlement persist despite digitalisation. (IMPRI India, 2025)
  • The Bridge Problem (Legacy Systems): Many departments still need to use older “legacy” data systems also, all datasets lack digital bridges, technically known as APIs. Without these bridges, two computers cannot talk to each other or share information in real-time. This forces officials to manually download, email, and re-upload data, a process that is slow and prone to errors.
  • Focusing on “What” instead of “Why”: Because data is trapped in separate administrative buildings, policies are often evaluated on “Outputs” (how many kilometres of road were paved) rather than “Outcomes” (did the new road actually reduce the time it takes for a farmer to reach the market?).

Key Systemic Gaps

The “Inter-Departmental Exchange” Gap

There is no standardized protocol to define what data is “shareable” versus “sensitive.” While the National Data Governance Framework Policy (NDGFP) first released by MeitY in 2022, seeks to address this by standardising data management across all government entities, it remains unenacted as of early 2026.

The Portability Bottleneck for Migrants

As of 2025, it is estimated that more than 600 million people have migrated within India’s borders, a substantial increase from the 450 million recorded in the 2011 Census, driven by ongoing urbanisation, recurring agrarian crises following the pandemic, and major shifts in the labour market.(Rajan and Nizam, 2025). Building on the success of the One Nation One Ration Card (ONORC)  which has made food entitlements portable across state lines , there remains an opportunity to extend similar portability principles to broader social security. As Srivastava (2020) notes, inter-state migrants holding BPL cards issued in their home state are often not considered for EWS housing schemes at their destination. The Code on Social Security 2020 presents an opportunity to address this gap by embedding portability and inter-state mobility as design principles across all centrally-sponsored schemes.

Jurisdictional Blind Spots in Urban Governance

The NITI Aayog’s report on urban planning capacity found that most states have not devolved planning functions to Urban Local Bodies as envisaged by the 74th Constitutional Amendment, resulting in several agencies operating at city and state levels with overlapping mandates ,creating accountability gaps and resource wastage (NITI Aayog, 2021).

Way Forward: MoSPI’s Roadmap for Data Harmonization and Data Linkage

Traditionally, MoSPI owns only specially curated sectoral survey datasets and administrative datasets are owned by programme Ministries/ Departments. However, to enable reuse of administrative datasets, and to  bridge the gap between fragmented systems and to build a responsive, citizen-centric government, MoSPI is leading a nationwide effort to align how departments manage and share data. As of 2026, this transition is being accelerated through four crisp initiatives designed to turn “data islands” into a shared national asset:

  • AI-Ready e-Sankhyiki Portal: In February 2026, MoSPI launched the Model Context Protocol (MCP) server for the e-Sankhyiki portal. Through initiatives such as the MCP server, MoSPI is enabling seamless integration of official datasets with AI tools and applications, making government data more accessible, interoperable, and usable for advanced analytics and evidence-based policymaking.
  • “Linkable-by-Design” Architecture: Datasets of Departments are often not discoverable because the datasets are not explained in a common language. ” MoSPI is fixing this by suggesting a standard framework, the National Metadata Structure (NMDS 2.0) for publishing referential metadata. It ensures data discoverability – the ability for systems to identify suitable datasets for a purpose  effortlessly without human intervention (MoSPI, 2026b).
  • A “Whole-of-Government” Asset: The April 2026 National Summit marked a shift from “data hoarding” to prepare harmonised shareable datasets by all State/UT Governments and agencies. A  “Whole-of-Government” approach for generating value by repurposing datasets. This Summit  acknowledged that data collected by one ministry is a vital resource for all. The focus can then be shifted to have  high-impact use cases- such as mapping water infrastructure spending against disease outbreaks to mathematically prove the public health return on investment.
  • Statistical Quality Assessment Framework (SQAF): For linked data to be effective, it must be statistically credible. The SQAF aims to bring in improvement in the National Statistical System by laying down quality parameters and corresponding good practices for statistical processes and products. By ensuring high-quality data at the point of entry, the government can slash the  analytical waste currently spent on fixing broken files, ensuring that when the system flags a family in need, the information is both timely and trustworthy.

Key Takeaways

  • The “Island” Problem: Government departments currently act like isolated islands. When data doesn’t travel between them, a citizen’s life is treated like a series of separate files rather than one whole story, leading to gaps where families can fall through.
  • The Missing Bridge: The key hurdle is not technology, but the absence of a common data language and the capacity-building necessary to effectively harness technology for secure and efficient data sharing.
  • Connection Over Collection: More data enables stronger evidence-based decision-making, but the real value emerges when data ecosystems are connected, interoperable, and effectively utilized.
  • Proactive, Not Reactive: The goal is a shift toward “Outcome-Based Governance.” By linking datasets, the government can move from fixing crises after they happen to providing proactive support, ensuring welfare is as mobile and dynamic as the citizens it serves. Also It can help make proactive policy interventions to improve governance impact.

 Conclusion

The transition from a “registration-focused” administration to an “outcome-focused” one is the next great frontier of the India Stack. With MoSPI’s push for shareable harmonised datasets in the context of AI-readiness  during the 2026 National Summit on Data Harmonization for Governance India can  move  towards a governance with a “Whole-of-Government” approach. It is time for India’s data to start speaking the same language for data governance. By weaving these “data islands” into a unified fabric, we can ensure that the promise of digital India is felt by every citizen at the very last mile.

About the Author

Mehakk Bhardwaaj is a Fellow with the Embark India Development Fellowship, placed at Administrative Statistics and Policy Division (ASPD) , Ministry of Statistics and Programme Implementation, Government of India. Her research values  Harmonisation and Interlinking of Data to improve governance efficiency.She is passionate about solving the last-mile delivery gaps in urban governance. All views are personal 

This blog was developed under the fellowship with mentorship from Ms. Jayasree M.G.( DDG , Ministry of Statistics and Programme Implementation), Dr. Samudra Sen (Assistant Professor & Program Director (MBA-GFM & AI&DS), RV University, Bengaluru) and  Ms. Pushpa (Research Fellow, GRAAM) 

References

 

 

Leave A Comment

About Us

Grassroots Research and Advocacy Movement (GRAAM) is a development research initiative in India focused on policy research, impact assessment, and strategic consultation. Collaborating with government, citizens, civil society, and corporate sectors, GRAAM ensures grassroots voices shape citizen-centric public policies. Their mission is to drive development by building human and social capital through evidence-based, community-informed solutions.

Popular Posts

Our Recent Posts

  • All Posts
  • Academics
  • Action research
  • Author
  • Challenges and Learnings in Skill Development
  • Community Development and Women Empowerment
  • CSR
  • Democratic Governance and Civic Participation
  • Development Research
  • Diversity Equity and Inclusion (DEI)
  • EIDF
  • Embark India Development Fellowship (EIDF)
  • Evaluating Digital Governance Initiatives
  • Evidence-Based Advocacy
  • Gig Worker
  • Global Conferences
  • Governance and Public Policy
  • Government Initiatives Evaluation
  • GRAAM Article
  • GRAAM Community Need Assesment
  • GRAAM Events
  • Graam in media
  • GRAAM Project
  • GRAAM Publication
  • GRAAM Research Mentorship
  • Grassroot Stories
  • Grassroots Initiatives
  • Lead Story
  • Media
  • Moments of the Month
  • News
  • NGO Resilience
  • Organizational Development
  • Organizational Milestones and Anniversaries
  • Photo clicks of the month
  • Policy Engagement
  • Research at GRAAM
  • Social business
  • Sugamya Shiksha
  • Sustainability and Environment
  • Uncategorized
  • Urban Governance
  • Workshop
  • Youth Empowerment Programs
  • Youth for Governance
    •   Back
    • CSR
    • Education
    • Livelihoods
    • Social Economy
    • Policy
    •   Back
    • Gig
    •   Back
    • Youth for Governance

Follow Us

Facebook Twitter Instagram Linkedin

Category

Tags

Edit Template
loader