Crowdmark — Canadian AI-Assisted Grading Platform Scaled Campus-Wide at the University of Waterloo

Crowdmark is an online assessment and grading platform founded in 2012 in Toronto by University of Toronto mathematics professor James Colliander and Martin Muñoz. It lets instructors scan or collect digital student work, then grade question-by-question across an entire class at once, reuse comments from a shared library, and export analytics on item-level performance.

The University of Waterloo piloted Crowdmark for roughly 12 months before running a public request-for-proposals process; it selected Crowdmark in July 2015 and now offers it free to instructors campus-wide, integrated with the university's learning management system and with Odyssey exam-logistics software. Waterloo's Centre for Teaching Excellence publishes its own guidance page recommending the tool and citing an approximate 50% reduction in grading time, alongside faster feedback and more consistent marking across teams of multiple teaching assistants.

Separately, Crowdmark's own published case study reports that grading teams at Northwestern University (21,000 students, departments including Economics, Mathematics, Physics, Chemistry and Psychology) saved approximately 16,660 hours of grading time, a roughly 40% reduction, alongside fewer regrade requests attributed to richer, faster feedback. These headline figures originate from vendor-published case studies and named faculty testimonials rather than an independently audited or peer-reviewed trial, so they should be read as illustrative rather than causally proven. Waterloo's own guidance is candid that adoption has real costs: paper exams still require scanning, teaching assistants need training to grade consistently, and digital submission tends to raise students' expectations for feedback detail.

Read the full analysis: https://uwaterloo.ca/crowdmark/