Bibliometric rankings have become proxies for research excellence in Indian academia. Critically examine their limitations and suggest a balanced framework for research evaluation.

Bibliometric rankings, such as the widely publicised "top 2% scientists" list, rank researchers by publication and citation counts. Indian universities now use them for hiring, promotion and branding. They are useful signals, but they distort research when treated as a final verdict on merit.

Why bibliometrics gained ground

• Objectivity and scale: uniform, public data reduce patronage in peer review and let a state-university researcher signal merit without a patron.
• National benchmarking: DST-CPR's scientometric study tracks India's output, citation impact and standing against G20/BRICS countries for 2014–2023 [1].

Limitations

• Goodhart's Law: once inclusion becomes a target, researchers chase citations through self-citation, citation cartels and honorary authorship.
• Quantity over quality: counts reward volume and visibility, not originality. The OECD notes that assessment still relies on narrow metrics that undervalue collaboration, openness and societal impact [2].
• Opaque data: the OECD calls rankings marketing tools built on non-transparent, proprietary data [2]. Errors in author profiles and discipline classification go unchecked.
• Linguistic exclusion: indexes dominated by English-language journals miss original work in Indian languages, which conflicts with NEP 2020. ANRF's planned SARAL AI will translate research into 18 Indian languages for outreach [3], but it does not change who gets credit.
• Disciplinary bias: humanities, agriculture and field sciences, where citation rates are low, are penalised. Long-gestation fundamental research is discouraged.

A balanced framework

• Metrics inform, experts decide: follow DORA and Leiden Manifesto principles and judge the content of research, not journal metrics. UNESCO's Open Science agenda supports this kind of assessment reform [4].
• Narrative CVs: ANRF grant reviews and UGC promotions should ask for a candidate's 3–5 best outputs with notes on what each contributed. Being on a list should never be a pass/fail filter.
• Wider impact: credit patents, policy inputs, teaching, open data and solutions to local problems.
• Open, inclusive indexing: build national databases on open, non-proprietary sources like OpenAlex, as the OECD suggests [2], and include Indian-language journals.
• Integrity safeguards: flag unusual self-citation patterns and curate journals to keep out predatory outlets.
• Evidence-based policy: rely on national studies like DST-CPR's [1] rather than commercial leaderboards.

Bibliometrics measure the symptoms of research activity, not its worth. A hybrid "metrics-plus-peer-review" model would keep their objectivity while restoring expert judgement. That model fits NEP 2020's vision of research-intensive, multilingual universities and SDG-4's goal of quality education, and it would let Indian science be judged on merit rather than labels.