Last week, I wrote about the daily misery of using an Indian court website: unreachable when you need it and unreliable when you reach it. It is tempting to conclude that maybe courts just need better designers and web developers. After all, they could hire the best in the country tomorrow and rebuild every website within a month. But that won’t change anything.
A court website is an access layer to an underlying data set that compiles case events — filings, listings, orders, and adjournments — that must be updated regularly with every case. Think of it as a window that lets you peek into a house. Painting the window or changing the interiors will only change how things look. It won’t fix the shaky foundation the house was built on.
Computerising the courtrooms
The eCourts project is nearly two decades old. It began in 2007 after the Supreme Court’s e-Committee prepared an action plan in 2005. That original plan was almost entirely about infrastructure. How to get courtroom computers, laptops, operating systems, Wi-Fi and training. There was little to no focus on data management — how a case should be identified, what forms part of its process record, how the provenance or the history of an entry is established, or who is responsible for maintaining its integrity.
The National Judicial Data Grid (NJDG) arrived in 2013, and with it came an attempt to standardise some of the reporting. It worked partially. Because of differences in practice and terminology, and sometimes laws, every state and district used its own nomenclature for case types. District courts recorded over 800 case types, which the NJDG sqeezed into around 30. Local difference is not by itself a problem. No one can or should force Kerala and Kashmir to speak the same language. But in the absence of translators and because no one has done the work of mapping these differences into a common structure, the reporting does not permit useful analysis.
The reasons for this are historical. The original unit of administration was the individual courtroom, treated as an entity unto itself. This is why in courtrooms across the country, there is a case number 1 for every case type, every year. It’s more like roll numbers in a classroom than a universal identity like an Aadhaar.
So now there are different practices and numbering systems, and different ways of recording interim applications (which are part of a case but are treated as separate cases entirely). And while there is a unique identifier, the Case Number Record or CNR, it’s almost never used in court documents.
If you ask questions like “What’s the most common cause for litigation in India?” or “What cases take up the most time across different courts” — there is no simple answer. We showed some of this in our previous work. There are issues with counting pendency, with court utilisation, and with understanding where the court spends time.
Also read: India’s court websites are broken. It’s becoming a justice crisis
Many cooks, no broth — big khichdi
All court websites carry a disclaimer. “Neither the Courts concerned nor the National Informatics Centre (NIC) nor the e-Committee is responsible for any data inaccuracy or delay in the updation of the data on this website. We do not accept any responsibility or liability for any damage or loss arising from the direct/indirect use of the information provided on the site.” So who really manages the data?
This absurd abdication of responsibility is symptomatic of the many issues that have crept up over two decades of court digitisation.
High courts don’t just have their own websites — as discussed in the previous article — but some, like Kerala, have built their own case management system. This is despite the fact that the feature list of CIS 4.0 — the centralised case management system set up by the e-Committee — looks really impressive. The intent was to create a national core with common functions, and a periphery that each state shall adapt to. But many courts have started complaining. It’s too rigid and too dependent on the NIC, which must also consider requests from all states.
Or let’s take e-filing. It became mandatory gradually in many places, often after much protest. Even today, stragglers exist. The Madras High Court’s blanket mandate for Tamil Nadu’s district courts was suspended in January 2026, after lawyers pointed to courts without scanners or bandwidth. Their annoyance is justified. An electronic filing does not flow into the Case Information System as the working court record. Lawyers are still asked for paper copies. Registries still maintain physical registers. We hear a lot of talk about digitisation, but any courtroom can make you experience the reality — lawyers and litigants arrive carrying large files and court staff shuffle tomes from one room to another. When a district court case reaches a high court, the lower court record is usually physically transported.
So it’s not surprising that lawyers don’t see the point of e-filing — why do additional work when the system eventually defaults to the old ways?
What’s the point of painting the pipes when the plumbing is broken?
Also read: We’re not ready for ‘One Case One Data’. Court records are too unorganised for digitisation
One way forward
A good data layer is the basis to any project of digital transformation. Technology is only useful when it changes not just the way information is accessed, but how that information is captured, managed and used for decision-making to make things better for the users — litigants, lawyers, judges and the court clerks. A good system must help administrators make better decisions instead of defaulting to the same instincts and old solutions — more money, more judges, more tribunals.
A good first step would be to define a core set of data standards — like what a case, a hearing, a listing, or an order actually means. There should also be some standard protocols to communicate information between two forums, as well as access to this through APIs providing machine-readable documents.
A dark mirror
One way to measure the success or failure of a public system is to look at the industry it has spawned. When India built dependable payment rails, the Digital Public Infrastructure of Aadhaar and UPI made payments easier, and an enormous fintech sector grew. Court data has produced a dark mirror of sorts: a legal-tech industry — Indian Kanoon, Provakil, CourtsDaily.info (disclosure: the author is one of the people behind the last one) — whose value-add lies substantially in cleaning, reconciling and re-serving data from official websites.
Imagine Swiggy having to name the streets, number the houses and build its own map before it could deliver food. Or Hyundai arriving in India and discovering that manufacturing cars came with a side business of repairing the roads.
What we need instead is the State providing reliable information and a robust data infrastructure. Different High Courts and district courts can, in turn, build applications on top of this — some in-house, some outsourced. These will allow for local configurability, multiple experiments and different innovations. They will solve for local priorities and problems and invite less resistance in adoption.
eCourts Phase III comes with a budget of Rs 7,210 crore, which was close to a billion dollars when it was approved in 2023. That money can go two ways: we can buy more paint for the windows and do some interior decoration, or we can choose to rebuild the foundation. At the very least, we can cement the cracks — data standards, definitions, APIs — so the systems that are built actually stand.
It’s time we rethink our approach to court digitisation.
Siddarth Raman is co-founder and CTO at The Professeer. Views are personal.
(Edited by Prashant Dixit)