Industry2026-09-02·8 min read

CBSE's OSM Rebuild for 2026-27: What the Board Is Doing Differently Before the Next Cycle

With a new chairman, an inquiry committee verdict, and binding Supreme Court directions, CBSE is redesigning its on-screen marking system for 2026-27 Class 12 examinations — here is what is changing and what it signals for the broader sector.

A Board Under Reconstruction

As of September 2026, the Central Board of Secondary Education is simultaneously running its compartment revaluation cycle, processing 1.6 lakh revaluation requests from the May Class 12 results, responding to the Supreme Court's August orders, and — quietly, in parallel — redesigning the on-screen marking system it intends to run for the 2026-27 Class 12 examination cycle.

The redesign is not publicly documented in a single announcement. It is visible through a sequence of administrative actions: the vendor inquiry, the new leadership's stated priorities, the Supreme Court's binding directions, and the NIT-audit model that emerged from the Radha Chauhan committee process. Together, these create a picture of what CBSE's 2026-27 OSM cycle is likely to look like — and what it signals for every examination body watching from the university sector.

What the Radha Chauhan Inquiry Found

The Cabinet Secretariat's one-member inquiry committee, chaired by S. Radha Chauhan of the Capacity Building Commission, was constituted in June 2026 with a specific mandate: examine the procurement and implementation of the OSM platform from Coempt EduTeck (formerly Globarena Technologies).

The committee's findings, shared with the Ministry of Education in August 2026, identified several specific failures in the CBSE procurement and deployment:

The lowest-bid problem: The OSM contract was awarded to Coempt as the lowest bidder, without adequate reference checks on prior deployments. Coempt's earlier work with the Telangana Board of Intermediate Education in 2019 had produced a controversy involving 3.8 lakh student failures and 21 suicides before results were corrected. This history was available and was not adequately factored into the procurement decision.

Scope without infrastructure verification: The contract scope covered 9.87 million answer scripts across approximately 70,000 evaluators — a deployment scale that the vendor had not previously handled. The board did not require proof of capacity before signing. There was no independent technical audit of the platform's ability to handle the volume.

Training as an afterthought: Evaluator training was provided through a video module and a brief orientation, not a structured competency-building program. The committee found no documented evaluation of whether evaluators had understood the interface or the marking scheme before evaluation began.

No pre-result quality gate: There was no systematic check before results were declared to verify that all booklets had been fully evaluated, that page-level scans were complete, and that mark totalling had been validated. Quality issues that might have been caught in a pre-declaration review instead became post-declaration complaints.

What CBSE Is Changing for 2026-27

CBSE's new chairperson, Lokhande Prashant Sitaram — a 2001-batch IAS officer with infrastructure project experience — has stated publicly that the 2026-27 examination cycle will incorporate specific safeguards that were absent in 2025-26. Based on statements from the board, Ministry of Education briefings, and the court-directed compliance roadmap, the changes fall into four categories.

Vendor Selection Reform

The 2026-27 OSM vendor selection will use a revised procurement framework that moves away from a pure lowest-bid model. Prospective vendors must now demonstrate:

Prior deployment at a scale of at least one million answer scripts with documented outcomes

A third-party technical audit of their scanning infrastructure conducted within the last 24 months

Reference checks from at least two prior clients, with independent verification, not just vendor-submitted certificates

A penalty clause tied to scan quality metrics — percentage of illegible scans, percentage of pages with tracking failures, and revaluation request rates attributable to platform issues

The NIT-audit model — where a National Institute of Technology or equivalent technical institution conducts an independent assessment of the OSM platform before deployment — is under active consideration as a mandatory pre-qualification step.

Technical Specifications for Scanning

The 2026-27 specification set, currently in draft form for vendor circulation, includes minimum technical requirements that were not part of the 2025-26 contract:

Page-level barcode/QR tracking: Every page of every answer booklet must carry a unique trackable identifier. Booklet-level tracking alone, which allowed page-order errors and missing-page failures in 2025-26, is no longer acceptable.

Minimum DPI for submission: Scans below 300 DPI will be automatically rejected by the platform. The scanning station must produce a per-booklet quality confirmation before the file is accepted into the evaluation queue.

Pre-evaluation completeness check: Before any booklet is released to an evaluator, the platform must confirm that all expected pages are present and that no page has a quality score below the threshold. Missing pages trigger a rescan workflow, not an evaluator notification.

Evaluator Training and Certification

The new training model, informed by the committee findings and the Supreme Court's specific observation about training documentation, moves from a video module to a structured three-step program:

Interface training: Evaluators complete a timed module on the marking interface, with a minimum passing score on a practice set. Completion is logged with a timestamp.

Marking scheme calibration: Each evaluator marks a calibration set of 20 answer scripts before the live evaluation begins. Calibration marks are compared against a gold standard. Evaluators outside the acceptable range receive targeted coaching before proceeding.

Live monitoring: During live evaluation, the platform tracks evaluator speed, marks-per-question distribution, and inter-evaluator consistency. Anomalies trigger a flag for chief examiner review.

All training records are maintained in a format that can be produced in response to RTI requests or court directions.

Pre-Declaration Quality Review

The most operationally significant change is the introduction of a mandatory pre-declaration quality review — a 48-72 hour window between the completion of evaluation and the uploading of results to the result system, during which a systematic automated check runs across the full evaluated dataset.

The check covers:

Completeness: every question slot in every booklet has a mark entry (zero or above)

Zero-mark detection: questions with a zero mark are flagged for human review unless the evaluator has recorded a specific reason code

Totalling accuracy: the sum of question marks matches the subject total in the evaluation record

Evaluator anomaly detection: scripts evaluated by flagged evaluators (statistical outliers) are pulled for secondary review

Only after this review clears can results be submitted to the declaration workflow.

What This Means for Universities

CBSE's OSM rebuild sets a de facto standard. When India's largest examination body — operating at 9.87 million scripts and 70,000 evaluators — specifies page-level QR tracking, pre-declaration quality gates, and structured evaluator certification, the implicit expectation for university-level OSM implementations rises alongside it.

More practically, university registrars and controllers of examinations who are currently evaluating digital evaluation systems should use the CBSE 2026-27 specification as a benchmark. If a vendor cannot demonstrate compliance with page-level tracking, pre-evaluation scan quality checking, and documented training workflows, they fall short of what the regulatory environment now expects — not aspirationally, but as a baseline.

The Radha Chauhan inquiry's findings about lowest-bid procurement without reference checks are a direct warning for institutional procurement processes. The pattern that produced the CBSE failure — contract awarded to a platform that had never operated at the required scale, with no independent audit, and no penalty clause for quality failures — is a pattern that institutional procurement committees can identify and avoid.

For institutions currently using paper-based evaluation and considering a transition, the CBSE rebuild also demonstrates one thing clearly: the path to a stable OSM deployment runs through pilots. CBSE skipped the pilot. PSEB Punjab did not. The outcomes reflect the difference. A phased rollout — one faculty, one semester, before full deployment — is not a delay; it is the design.

The CBSE Compartment Exam: An Early Test

CBSE's OSM compartment examination in July 2026 served as a limited live test of the post-crisis changes. The compartment cycle covered a substantially smaller volume than the main examinations, but new scanning protocols and a revised quality check process were applied.

The compartment results, declared in August, produced a fraction of the complaints volume from May. CBSE's new leadership described the compartment cycle as a "calibration exercise" and published a summary of the quality metrics — an unprecedented level of transparency for the board. The implication is that similar metric publication is likely for the 2026-27 main cycle.