Annoto vs. Perusall: Social Annotation for Readings, or for Video?
Perusall does one thing well: students read a document together, highlight it, ask questions in the margin, and get an engagement score for the effort. It was built around the textbook, and the workflow still shows it.
Annoto brings the same social instinct to the lecture video: graded questions on the timeline, replies pinned to the second, video assignments, peer review, and comprehension data you can act on.
Learn more: what changes when social learning moves from the page to the video
From Annotating the Reading to Running the Video Course
A social reading platform and a video engagement layer look alike in a demo: both anchor comments to a point in the material and both score participation. They diverge over a semester, in three places: what the material is, where it has to live, and what evidence the institution can produce.
What is Perusall built for?
Perusall targets the pre class reading. You upload a document or assign a textbook from its catalogue, students highlight and comment in the margin, and an engagement algorithm scores how much, how well and how early they annotated. A confusion report surfaces the passages that drew the most questions. Videos and podcasts can be assigned too, but they are items in the same library rather than the centre of the design.
Where does higher education need more?
Lecture capture, demonstrations and recorded seminars are where most course hours now sit, and they need the same social layer the reading gets. That is what timestamped discussion gives you. Assessment has to reach past annotation counts, through peer review, video assignments, and quizzes on the timeline that Lumo AI can draft from the lecture itself. And the institution needs evidence it can hand to a reviewer: attention and comprehension per learner and per moment, exportable for programme review and Regular and Substantive Interaction reporting.
Where does the video live?
Perusall keeps content in its own library: you upload the file or paste a link and students open it inside Perusall. Annoto overlays the player you already run, whether that is Panopto, Kaltura, YouTube or Vimeo. It also hosts and streams video natively, so a department without a video platform can run on Annoto alone. Your library keeps its permissions and its captioning pipeline either way, and nobody re uploads a lecture to annotate it.
How does Annoto install?
Two routes, and you pick by where the video sits. Inside an LMS, an admin registers Annoto as an LTI 1.1 or 1.3 tool, and sign-on, rosters and grades travel through the standard. Anywhere else, you add one script tag and a client ID. The JavaScript SDK attaches the layer to your player, and a REST API and webhooks move interaction data into the systems you already run. Custom portals build on the same SDK that carries Annoto's own integrations.
What does an instructor get back for the setup time?
One pass over a lecture produces graded questions, discussion prompts, reflection points, and a comprehension baseline for the cohort. Later terms inherit all of it, because the layer attaches to the video rather than to a copy of it.
Which one should we choose?
A course built around assigned readings and a shared textbook will be well served by Perusall. A course built around lecture video, demonstrations or student recordings needs discussion, assessment and analytics on the timeline, and that is the case for a video engagement layer. Many departments run both: Perusall for the reading, Annoto for the video, each syncing to the same gradebook.
What does LTI 1.3 change for the people who own the LMS?
Perusall connects to the LMS over LTI and syncs grades back, but each launch opens the Perusall platform and students do their work there. An LTI 1.3 registration of Annoto works the other way round. An administrator registers Annoto once at the platform level, every course in the tenant inherits the tool, and the engagement layer renders inside the LMS page on the LMS video, so nobody leaves the course shell. Identity, role and roster arrive with each launch, grades return over the same registration, and IT reviews one tool, rotates one key and revokes access in one place. See compliance for the standards involved.
How does a department roll this out without a migration project?
Perusall asks you to upload or link each item into its library before anyone can annotate it. Annoto attaches to the video rather than to a copy of it, so the deployment question stays narrow: where does the player live, and who owns the library? Media teams keep the Panopto, Kaltura, YouTube or Vimeo pipeline they already run, with its permissions, retention rules and captioning workflow, and Annoto renders over it. A department with no video platform publishes into Annoto's native hosting instead. Neither route re encodes a back catalogue or leaves a second copy to keep in sync, and recordings reused next term carry their questions and discussion.
Why is assessment the harder requirement?
Perusall grades participation: an algorithm weighs how much students annotated, how thoughtful the comments look, and how evenly they are spread through the reading, and its peer review assignments add rubrics for written work. That serves reading compliance and essay feedback well. Universities also assess presentations, clinical technique, teaching practice and studio work, where the artefact is a video and the judgement needs a rubric applied to what happens on screen. Video assignments make the submission a video with a timeline, instructors and peers score it against assessment criteria, and feedback lands as timestamped comments rather than a paragraph at the bottom of a form. Teaching assistants moderate in place.
What does a programme review need that an engagement score cannot give?
An engagement score says who annotated enough. A programme review asks which minutes the cohort rewatched, where attention fell away before an assessment, how participation is distributed rather than averaged, and which learners disengaged early enough to reach. Annoto's analytics report attention and comprehension per learner and per moment, and export by CSV or API, so the evidence sits beside your institutional reporting. That is the textbook inheritance showing: a reading platform answers whether students did the reading; a university answers to course teams, quality reviewers and accreditors about the whole course.
Annoto
