TL;DR: Indian Kanoon is a free, enormous, well-indexed database of Indian judgments and statutes, and it is where most Indian lawyers correctly start a search. It stops short of being a full research tool in six specific ways: no top-of-case good-law flag, no human-edited headnote, uneven coverage across smaller High Courts and district courts, no synthesised view of how later courts treated a precedent, a document ID that is not a citation format most courts expect, and full-text search that returns every mention of your terms rather than the paragraph carrying the ratio. Each gap has a workaround, covered below.

On this page

Why this comparison matters

Indian Kanoon holds a working monopoly on the first click of Indian legal research. It is free, it does not gate the full text of a judgment behind a paywall, and its search box returns results fast. For a law student checking a proposition, a junior associate scanning for prior art on a point, or a self-represented litigant trying to understand a section, it is usually the right first stop.

The site is also honest about what it offers. Its own free features page lists search, cross-linking between documents, a machine-generated structural breakdown of long judgments, and a machine-generated sentiment tag on cited text. None of that is a hidden shortfall. The gap between what a lawyer needs before filing and what a free full-text index can supply is structural, not a defect in Indian Kanoon’s execution. This piece maps that gap precisely, states how each claim below was checked against the live site on 2 August 2026, and gives a workaround for each one so a lawyer who keeps using Indian Kanoon, which most will, can use it safely.

For background on why any of this matters, see what good law means and how to check it and how to cite Indian judgments. This post assumes you have read those and goes further into where the free tool stops.

What Indian Kanoon actually does well

Before the limitations, the parts that hold up under scrutiny, each confirmed by browsing the live site.

The database itself is broad. The site’s own browse index lists all 25 High Courts, the Supreme Court, 19 tribunals including the Income Tax Appellate Tribunal, the National Green Tribunal, and the Competition Commission of India, plus Constituent Assembly Debates and Law Commission reports. A search for a common criminal provision such as Section 138 of the Negotiable Instruments Act returns results in the thousands within a second, filterable by court, year, judge, and document type.

Cross-linking is real and functions as advertised. Every judgment on the site displays a “Cites” and a “Cited by” count, each hyperlinked to the actual list of documents. On the Supreme Court’s Shreya Singhal v Union of India judgment, the page shows the case cites 111 other decisions and is cited by 303 later ones, each one a working link to the source text, not a dead reference.

The site also documents two machine-learned add-ons that most users are not aware of: a “Structural Analysis” table that segments judgments longer than five paragraphs into Facts, Issues, Arguments, Precedent Analysis, Analysis of Law, Court’s Reasoning, and Conclusion, and a “Reliability of a Precedent” tool that tags individual cited passages as Relied by Party, Accepted by Court, Negatively viewed by Court, or No clear sentiment. Both are listed as free features on Indian Kanoon’s own page. Neither is a substitute for the things below, but they exist and a researcher who has never opened the left sidebar on a judgment page is missing free tooling.

None of this is Autoposting-style marketing. It means the honest limitations are narrower and more specific than “the free tool is basic.” They sit in six places.

Limitation 1: no good-law flag at the top of a judgment

Open any judgment page on Indian Kanoon, including a landmark one that has since been partially overruled or read down, and there is no badge, banner, or colour code near the case title telling you its current standing. The page for ADM Jabalpur v Shivkant Shukla, a 1976 judgment substantially overruled by the nine-judge bench in K.S. Puttaswamy v Union of India (2017) 10 SCC 1, carries no notice of that fact on the document itself. You have to already know the overruling happened, or find it through your own search.

This was confirmed by opening the Supreme Court judgment page for Shreya Singhal directly (document ID 110813550) and checking the page structure at the top: citation block, bench, then the judgment text. No status indicator sits anywhere in that block. Commercial citators built for this exact purpose, SCC Online’s citator and Manupatra’s KLI, place a synthesised status flag (good law, partly overruled, doubted) at the top of a judgment record, built from editorial review of every later citing case. Indian Kanoon’s per-passage sentiment tag, described above, is scoped to how one citing document treated the case in that one instance. It is not the same thing as a synthesised, current-status verdict, and it is not positioned as one.

Workaround. Treat a citation count as a lead, never a conclusion. Before relying on any precedent more than a few years old, open the “Cited by” list, sort by most recent, and read the last five to ten citing judgments for how they characterised the earlier case. The full method, including a worked example on ADM Jabalpur itself, is in good-law checking. A citator that surfaces overruled or doubted status as a flag rather than requiring a manual read of the citing list is what a captive good-law tool adds on top of this workflow.

Limitation 2: no human-edited headnote

Reporter series like SCC and AIR employ legal editors who write a headnote: a short, professionally drafted summary of the holding, cross-referenced to the paragraphs that support it. Indian Kanoon does not do this. What it offers instead, per its own free_features page, is a machine-learned structural table for documents over five paragraphs, marking sections as Facts, Issues, Arguments, and so on.

This is useful for navigation inside a long judgment. It is not the same product as a headnote. A structural label tells you which paragraphs are “the court’s reasoning”; it does not tell you what the court actually held, phrased as a rule you can cite. Two judgments can both have a “Court’s Reasoning” section spanning fifteen paragraphs, one of which announces a bright-line rule and the other of which weighs six factors without settling on one. The structural tag treats them identically. A headnote written by a legal editor would not.

Workaround. Read the operative paragraphs yourself rather than relying on any summary, machine or editorial, as a substitute for the primary text. When you need a fast first pass before deciding whether a judgment is worth reading in full, how to read a judgment sets out where the ratio typically sits in an Indian judgment’s structure, so you know which paragraphs to go to directly rather than skimming the whole text or trusting an auto-generated label.

Limitation 3: uneven High Court and district court coverage

Coverage is not uniform, and the unevenness is not where instinct suggests it would be. Checking Indian Kanoon’s Manipur High Court browse page directly on 2 August 2026 shows the Manipur High Court’s year index jumps from 1800 (102 documents, almost certainly Privy Council-era material catalogued under the modern court) to 2013 (a single document), then to a complete gap from 2014 through 2019, before resuming at 2020 with 5,835 documents and climbing sharply each year after. Six consecutive years with zero indexed judgments for a functioning High Court is a real hole, not a rounding artefact.

By contrast, the Sikkim High Court’s browse index runs continuously from 1977 through 2026 with no comparable gap, disproving an initial assumption that smaller northeastern High Courts are uniformly under-indexed. The pattern is court-specific, not size-specific, and has to be checked per court rather than assumed.

District court coverage is the sharper limitation. The site’s browse index lists exactly two district court jurisdictions: Delhi District Court and Bangalore District Court (confirmed live on the browse index). India has several hundred district court complexes. If your matter sits in a district court outside Delhi or Bengaluru, and a large share of Indian litigation does, Indian Kanoon simply has no index for it. This is not a search-quality problem to route around with better keywords; the underlying documents are not there.

Workaround. For a High Court matter, check the specific court’s own browse page for gaps before treating an absence of results as an absence of precedent, especially for anything before 2020. For a district court matter outside Delhi or Bangalore, go directly to the eCourts case status and NJDG portal, the live National Judicial Data Grid itself, or the relevant High Court’s own e-filing and cause list system, since Indian Kanoon is not the index of record there.

Limitation 4: cited by counts without synthesised treatment

The “Cited by” number on a judgment page is accurate and the underlying list is real, confirmed by opening the list behind Shreya Singhal’s 303 citing cases. What it does not give you is a synthesis. A commercial citator would tell you, in one line, how a body of 303 later judgments collectively treated the precedent: still followed, distinguished in a specific fact pattern, doubted by a coordinate bench, or referred to a larger bench. Indian Kanoon gives you the raw list and, within each citing document, a per-passage sentiment tag if the citing text was substantial enough to trigger one. It does not roll that up into one answer at the case level.

For a judgment cited hundreds of times, reading each one is not realistic within a filing deadline. For a judgment cited a handful of times, it is entirely feasible and often the more reliable approach anyway, since it puts you in the actual language of the later court rather than a third party’s characterisation of it.

Workaround. Sort “Cited by” by most recent first. For a heavily cited case, read the ten to fifteen most recent citations rather than trying to cover all of them; later courts are the ones most likely to have addressed a subsequent overruling or a conflicting larger-bench reference. For a lightly cited case, read all of them. Either way, you are reading primary text, which is also the safer habit given how often AI summarisation tools misstate case treatment, a problem covered in the lawyer’s duty to verify AI output.

A citator layer that pre-reads the citing list and flags subsequent negative treatment removes the manual step for the heavily cited cases, which is exactly where manual review is least practical. That is the specific gap a product like Niyam’s good-law citator is built to close: it does not replace reading the primary text, it tells you which of the hundreds of citing cases are worth reading first.

Limitation 5: a document ID is not a citation

Every Indian Kanoon judgment lives at a URL like indiankanoon.org/doc/110813550/. That number is a stable, permanent identifier for the document on that site, and it is genuinely useful for sharing a link to a colleague or bookmarking a case. It is not a citation format that a court, a senior counsel settling a brief, or an opposing party’s lawyer will recognise or accept in a pleading.

Indian courts expect neutral citations issued by the court itself (2024 INSC 835 format for the Supreme Court of India since 2023, or a High Court’s own neutral citation such as 2023:DHC:2720), or a reporter citation from AIR, SCC, or SCR. Indian Kanoon does display these citations when they exist, usually at the top of the judgment page alongside the document ID, confirmed on the Shreya Singhal page, which shows AIR 2015 SUPREME COURT 1523 and 2015 (5) SCC 1 above the judgment text. The limitation is not that the citation is missing; it is that the document ID is the more prominent, easier-to-copy reference, and it is easy for a researcher in a hurry to cite the URL or the doc number instead of the actual citation sitting a few lines above it.

Workaround. Never cite an indiankanoon.org URL or a bare document ID in a filing. Copy the neutral citation or reporter citation displayed at the top of the page instead, and verify it independently against the Supreme Court’s own e-SCR portal or the eCourts judgments search before use, since even correctly formatted citations can be typed incorrectly on any site. The full mechanics of citation families, brackets, and pinpoint references are in how to cite Indian judgments, and the specifics of the post-2023 neutral citation scheme are in neutral citations and the e-SCR.

Limitation 6: full-text search surfaces mentions, not the ratio

Indian Kanoon’s search is a full-text index. Searching “cheque bounce section 138” returns roughly 9,485 results, confirmed live, ranked by a relevance algorithm that weighs term frequency and proximity. This means a judgment where Section 138 is discussed in passing, cited once in a list of provisions considered and never analysed again, can rank near a judgment that turns entirely on the interpretation of Section 138. The search engine does not distinguish between a section being the subject of the judgment and a section being mentioned in it.

This is not a flaw unique to Indian Kanoon. Every full-text legal search engine, free or paid, has this property to some degree, because full-text search ranks on word statistics, not on legal significance. The AI Tags feature visible on Indian Kanoon’s search results page (confirmed on the Section 138 search, showing tags like “cheque-dishonour” and “presumptions”) narrows results by machine-classified topic and helps, but does not solve the underlying issue: a topic tag tells you the document is about the general subject, not that the specific paragraph you need sits inside it.

Workaround. Treat the first page of full-text results as a shortlist, not an answer. Use the court, year, and AI tag filters to narrow before reading, then open each shortlisted result and use your browser’s find-in-page to jump to the actual paragraph where your search term appears, checking whether it is analysed or merely listed. For the difference between a source that states the rule (a judgment, a statute) and a source that discusses the rule (a commentary, a headnote, a case digest), see primary vs secondary legal sources, since Indian Kanoon’s search ranking does not distinguish between the two either.

What Indian Kanoon does well vs what it does not do

CapabilityIndian KanoonWorkaround
Free full-text access to judgments and statutes✓None needed
Cross-linked cites and cited-by counts✓None needed
Coverage of all High Courts and major tribunals✓Check per-court browse page for date gaps before assuming completeness
Machine-generated structural breakdown of long judgments✓Read the reasoning paragraphs directly, do not treat the label as the holding
Good-law status flag on a judgment✗Read the most recent citing cases yourself, or use a citator layer
Human-edited headnote✗Read the operative paragraphs; see how to read a judgment
District court coverage beyond Delhi and Bangalore✗Use eCourts and NJDG or the relevant court’s own portal
Synthesised subsequent treatment across all citing cases✗Sort cited-by most recent first and read the last ten to fifteen
Citation ready for a pleading, by default✗Copy the AIR/SCC/neutral citation shown on the page, not the doc ID or URL
Search that ranks by legal significance rather than word frequency✗Filter by court, year, and AI tag first, then read for the ratio

A verification workflow that starts on Indian Kanoon

The limitations above point to one habit: use Indian Kanoon to find candidate authority, then verify before you rely on it. The workflow looks the same whether the underlying question is about good-law status, citation accuracy, or coverage gaps.

flowchart TD
    A[Search Indian Kanoon for the proposition] --> B[Shortlist candidate judgments]
    B --> C{Court and district covered?}
    C -->|No, e.g. district court outside Delhi/Bangalore| D[Go to eCourts / NJDG or the court portal]
    C -->|Yes| E[Open the judgment, read the operative paragraphs]
    E --> F[Note the citation shown on the page: AIR/SCC/neutral]
    F --> G[Open Cited by, sort most recent first]
    G --> H{Any negative treatment in recent citations?}
    H -->|Yes| I[Read those citing judgments in full before relying on the case]
    H -->|No signal found| J[Cross-check citation on e-SCR or the reporter itself]
    D --> J
    I --> J
    J --> K[Cite the verified reporter or neutral citation, not the doc ID]

Each node in that flow is a step already described above. The point of drawing it out is that none of the six limitations requires abandoning Indian Kanoon; each requires one additional, specific check before the result goes into a filing.

A worked example: tracing one search to a citable source

Take a junior associate researching whether a cheque-bounce complaint under Section 138 of the Negotiable Instruments Act, 1881 (now read alongside BNSS provisions replacing the old CrPC procedure) survives a limitation objection. A search for “cheque bounce section 138 limitation” on Indian Kanoon returns thousands of results ranked by relevance, several of which discuss limitation only as one issue among many.

The associate filters by court (Supreme Court, to get binding authority first), sorts by most recent, and finds a candidate judgment. Reading the operative paragraphs directly, rather than trusting a snippet, confirms the judgment actually decides the limitation point rather than mentioning it in passing. The page displays a reporter citation, say, a 2023 SCC reference, above the text. The associate opens “Cited by,” sorted most recent first, and checks the last ten citing judgments for any indication the point was later doubted or distinguished on similar facts. Finding none, the associate cross-checks the SCC citation against the e-SCR or the reporter itself, then cites that verified reference in the draft, never the indiankanoon.org document ID.

That is six extra minutes of work on top of the search itself. It is also the difference between a citation a senior counsel will accept without a second look and one that gets flagged in review.

Where a citator layer helps

None of the workarounds above are exotic; they are what careful researchers already do by habit. What they cost is time, specifically the time spent manually reading the last ten citing cases on every precedent worth relying on, across every matter, every week. A citator built to flag subsequent negative treatment automatically, of the kind good law checking describes in detail, does not replace the primary-text reading. It tells you which of the citing cases are worth reading first, which is the part that does not scale by hand once a precedent has been cited a few hundred times. Niyam’s citator applies that check across a search result set rather than one judgment at a time, which is the specific gap between “Indian Kanoon found the case” and “the case is safe to file with.”

The same principle applies to search itself. Indian Kanoon’s full-text ranking surfaces mentions; a search layer built to weight results by whether a provision or proposition is actually the subject of the judgment, not merely present in it, narrows the shortlist before a human has to open every result. That is the problem good AI legal research tools are built to sit on top of, rather than replace, a free full-text index.

A checklist for using Indian Kanoon safely

  1. Treat every search result as a shortlist, not an answer; filter by court, year, and topic tag before reading.
  2. For any case older than five years, or from a smaller High Court, check the court’s browse index for date gaps before assuming full coverage.
  3. For a district court matter outside Delhi or Bangalore, do not expect Indian Kanoon to have it; use eCourts or the court’s own portal.
  4. Read the operative paragraphs of a judgment yourself; do not rely on the structural table label as a substitute for the holding.
  5. Sort “Cited by” most recent first and read the last ten to fifteen citing judgments before relying on any frequently cited precedent.
  6. Copy the reporter or neutral citation shown on the page, never the document ID or the URL, into any filing.
  7. Cross-check the citation against the e-SCR or the reporter series before the citation goes into a final draft.

Frequently asked questions

Is Indian Kanoon free to use?

Yes, the core database, search, advanced search, and cross-linking between documents are free, confirmed on indiankanoon.org/free_features/. The site sells a paid tier called Prism AI for research assistants, drafting tools, and higher download limits, but access to judgment and statute text itself is not paywalled.

Does Indian Kanoon tell you if a judgment has been overruled?

Not with a status flag on the case page itself. It shows a “Cited by” count and list, and within citing documents a machine-generated sentiment tag on specific cited passages, but it does not display a synthesised current-status indicator like “overruled” or “good law” at the top of a judgment. You have to check the citing cases yourself or use a dedicated citator.

Can I cite an Indian Kanoon URL in a court filing?

No. Courts expect a reporter citation (AIR, SCC, SCR) or, for judgments after the scheme’s rollout, a neutral citation such as 2024 INSC 835. Indian Kanoon usually displays these at the top of the judgment page above the document text; copy that, not the indiankanoon.org/doc/ URL or ID number.

Does Indian Kanoon cover all High Courts?

It indexes all 25 High Courts through its browse index, but coverage depth and date ranges vary by court and are not uniform by size. Checking the Manipur High Court browse page on 2 August 2026 showed a complete gap in indexed judgments from 2014 through 2019, while the Sikkim High Court’s index ran continuously across the same years. Check the specific court’s browse page before assuming completeness.

Does Indian Kanoon cover district courts?

Only two: Delhi District Court and Bangalore District Court, confirmed via the site’s own browse index. For any other district court in India, Indian Kanoon has no matching coverage; use the eCourts and NJDG portal or the relevant court’s own case information system instead.

What is the “Reliability of a Precedent” feature on Indian Kanoon?

It is a free, machine-learned feature described on the site’s own features page that tags passages where a judgment cites an earlier case as Relied by Party, Accepted by Court, Negatively viewed by Court, or No clear sentiment. It operates per citing passage, inside one document, rather than as a synthesised status for the cited precedent as a whole, so it is not the same as a top-of-case good-law flag.

Does Indian Kanoon have headnotes like SCC or AIR?

No editorial headnotes written by legal editors. It offers a machine-generated structural table on judgments over five paragraphs that labels sections as Facts, Issues, Arguments, Reasoning, and Conclusion, which helps with navigation but does not state the holding the way a professionally drafted headnote does.

Why does a keyword search on Indian Kanoon return irrelevant results?

Indian Kanoon’s search is a full-text index that ranks by word frequency and proximity, not by legal significance. A judgment that mentions your search term once in a list of provisions considered can rank alongside one where the term is the actual subject of the decision. Filter by court, year, and the AI tags shown on the results page, then read the operative paragraphs to confirm relevance.

Is Indian Kanoon’s citation count for a judgment accurate?

The “Cites” and “Cited by” counts reflect real, hyperlinked documents rather than placeholder numbers, confirmed by opening the citing list behind a Supreme Court judgment directly. What the count does not give you is a summary of how those citing cases treated the precedent; you have to open the list and read it, or rely on a per-passage sentiment tag inside each citing document.

Should a law student rely on Indian Kanoon alone for research?

For learning how a provision has been interpreted and for finding candidate authority quickly, yes, it is a strong free starting point. For anything going into a moot memorial, an assignment requiring formal citation, or later professional practice, cross-check the citation format and read the primary text rather than relying on search snippets or the structural summary alone.

How current is Indian Kanoon’s Supreme Court coverage?

Very current for the Supreme Court specifically; recent Supreme Court judgments typically appear within days through the site’s daily orders feed. The unevenness described in this piece is concentrated in smaller High Courts and, more sharply, in district courts, where coverage is limited to Delhi and Bangalore.

Does Indian Kanoon show which bench or judge decided a case?

Yes. Author and bench information is displayed on the judgment page and is also filterable in advanced search, including a “judges resolution” feature that has identified over 2,200 judges, covering all Supreme Court judges and High Court judges appointed after 2008.

What is the difference between Indian Kanoon and a paid citator like SCC Online or Manupatra?

The core difference is editorial synthesis. Paid citators employ legal editors who review how later courts treated a precedent and publish a single, current status (followed, distinguished, overruled, doubted) at the case level. Indian Kanoon gives you the raw citing list and a machine tag on individual passages, leaving the synthesis step to the researcher.

Can AI tools built on top of Indian Kanoon fix these limitations?

Partially, depending on what the tool is built to do. A tool that reads the citing list and flags negative subsequent treatment addresses limitation 4. A tool that ranks search results by whether a provision is the actual subject of a judgment, not merely mentioned, addresses limitation 6. No tool built on Indian Kanoon’s data can fix limitation 3, the coverage gaps, since the underlying documents for many district courts are not in the index to begin with.

Where should I check a citation Indian Kanoon shows me before filing it?

Cross-check against the Supreme Court’s own e-SCR portal for Supreme Court judgments, the relevant High Court’s official judgment portal for High Court decisions, or the reporter series itself (AIR, SCC) if you have access. See neutral citations and the e-SCR for the verification steps.

Does Indian Kanoon index gazette notifications and circulars?

The browse index covers judgments, statutes (Central Acts and state legislation), tribunal orders, and parliamentary debates, not gazette notifications or regulatory circulars as a distinct, current category. For the authoritative text of a Central Act, cross-check against India Code, the government’s own statute repository; for gazette notifications, the official gazette and the issuing regulator’s own website remain the primary source.

Indian Kanoon’s own limitations are not a reason to stop using it first. They are a reason to stop treating the first search result as the final answer. The six gaps above, no good-law flag, no editorial headnote, uneven coverage below the High Court level, unsynthesised citation counts, a document ID that is not a citation, and relevance-ranked rather than significance-ranked search, are each narrow enough to work around with one additional, specific check. The workaround is usually a few extra minutes; skipping it is the part that shows up in a court’s order.