Perspective 1Design + AI Mindset

The Citizen Should Not Be the System Integrator

The daily friction of Indian life, from traffic to clunky portals, is not bad luck. It happens when people are left to connect rules, counters and handoffs that nobody connected for them.

Every step done, nothing achieved

The story below is illustrative, built from common service journeys. It is not about a real person.

A woman needs to change the name on her family's electricity connection after her father's death. The portal accepts her application. Then it asks for a document from another department. That department needs a certificate from a third office, which asks her to visit in person. Her application on the first portal times out while she waits.

She did everything she was asked. Each office did its part. Nobody was responsible for whether she reached the end.

This is what we mean by an invisible flaw. It is not a broken server or a rude clerk. It is a system in which the citizen has become the integrator, carrying information and documents between institutions that were never designed to work as one.

More is in place than most people realise

India's digital public infrastructure is real. As of 31 July 2026, the UMANG app offered about 2,575 government services and DigiLocker offered 5,437 document types and services. States and union territories reported 24,563 e-services, and 1,735 of 2,124 mandatory e-services were available online, according to the national e-governance service delivery assessment of January 2026.

Several states have unified portals that offer all 59 mandatory services. The problem is not a lack of technology or effort.

Where the friction hides

  • Finding the service. People often do not know which office, portal or scheme handles their situation.
  • Proving the same thing again. The same documents are requested by department after department.
  • Invisible handoffs. When an application moves between offices, the citizen cannot see where it is or who owns it.
  • Exceptions become dead ends. The hardest 10% to 20% of cases, a name mismatch or a missing record, usually end in an office visit.
  • Counting the wrong thing. Success is measured by applications processed, not by whether people got what they needed.

Each actor can be doing their job well while the end result for the citizen stays poor. That is the signature of a design problem.

Why design, and why AI

Design, in the sense we use it, is not decoration. It is the discipline of deciding how a service should work for the person using it, end to end, and testing whether it does. Countries such as the United Kingdom, with its government service standard, and Estonia, with its "once-only" rule that the state should not ask for the same information twice, have shown what this looks like at national scale.

AI matters because India's scale is unlike anywhere else. A design that works in one district has to reach hundreds, in many languages. AI can cluster the reasons applications fail, summarise a case history for the next officer, flag journeys where people repeatedly get stuck, and explain the next step in plain language. Decisions about rights and eligibility must stay with an authorised person, with a way to appeal.

The redesign: make the outcome the unit of service

A design to test. Not a claim of achieved outcomes.

We propose five building blocks that any department can adopt:

  1. An outcome journey map.

    Before any software is bought, map the whole journey, from trigger to outcome, with a named owner for the end result.

  2. Once-only evidence.

    If one part of government already holds a document, no other part should ask the citizen for it again.

  3. One case reference.

    A single number that follows the citizen across departments, so anyone can see where the case is.

  4. An exception lane.

    A defined route, with an owner and a deadline, for the cases that do not fit the standard path.

  5. A citizen effort score.

    Measure visits, documents and days needed to complete a service, and publish it.

Start with one service

Pick one high-volume service, such as an electricity name change or an income certificate, in one district. Run it for 90 days with 500 to 1,000 cases, first with people and simple tools, then with only the digital pieces that prove their worth. Our planning estimate is ₹20 to 35 lakh.

The measures that matter are first-time completion, visits avoided, documents removed, median completion time and repeat contacts. Downloads and dashboards are not outcomes.

What each of us can do

Who can act and the next step
Citizenswhen a service fails you, notice the exact step where it broke. That is design research.
Department headsname one person accountable for the whole journey of your most-used service.
State digital teamspublish completion time and citizen effort, not just the number of services online.
Founderslook for the gaps between two systems that already exist. That is where the most useful products are.
Students and designerslearn to map services, not only screens.

Design is how a better India gets built

India does not need more portals for their own sake. It needs services that work the first time, for the person using them. That starts when someone owns the whole journey. Design + AI = Better India.

Watch the full episode →

Redesigning India: The Invisible Flaws Holding Us Back

Your turn: what is the one everyday friction in India you most want us to redesign? Tell us in the comments on the video, and subscribe to Better India By Design for the next redesign.

Watch on YouTube →

Sources and notes

Figures reflect the periods, populations and definitions in the cited material. They are not live totals.

  1. 01
    Press Information Bureau, "Government Expands Citizen-Centric Digital Public Service Delivery through UMANG and DigiLocker", 2026. Link
  2. 02
    Press Information Bureau, "NeSDA Way Forward", January 2026. Link

Pilot costs are Better India By Design planning estimates, not government figures.

Researched with AI assistance. Facts checked against the cited sources and reviewed by the BIBD Editorial Team.

BIBD proposals are designs to test, not claims of achieved outcomes. Pilot costs are planning estimates unless a source states otherwise.

Website edition prepared from the supplied Perspective on 31 October 2026. Original document →

Continue reading

Another system. A connected question.