What is a video abstract? A guide to research videos
By ShinobiMedia.ca · Video · Published
A video abstract is a short video, usually one to three minutes, that explains what a research paper asked, how the study was done and what it found, for people who will not read the paper. This guide covers what a good one contains, how to script it from a manuscript, how it differs from the other kinds of research video, what Canadian funders mean by knowledge mobilization, and the accessibility requirements that apply to video published by Ontario institutions.
Get a free quoteCall (613) 800-8028
What a video abstract is, and what it is not
It is a summary of one piece of research for a viewer who has not read it. That viewer might be a researcher in a neighbouring field deciding whether the paper is relevant, a clinician, a policy analyst, a journalist or a member of the community the study involved.
It is not a recorded conference talk, a lecture or a promotional film about a lab. Those have their uses, but they assume a viewer who has already decided to pay attention. A video abstract has to earn that in its first ten seconds.
The kinds of research video, compared
| Format | Typical length | Made for |
|---|---|---|
| Video abstract | 1 to 3 minutes | One paper, for its readers |
| Plain-language explainer | 1 to 2 minutes | Patients and the public |
| Project or lab overview | 2 to 4 minutes | Funders and partners |
| Report summary | 2 to 3 minutes | Policy staff and boards |
| Social cut-down | 20 to 40 seconds | One result, in a feed |
| Poster or conference loop | 30 to 90 seconds | A screen beside a poster |
Lengths are working conventions, not rules. A journal's own author guidelines override them.
How to structure the script
Write the script before anything else, and write it for the ear. A sentence that reads well in an abstract is often too dense to follow when spoken once at speed.
- The problem, in one or two sentences a non-specialist would recognise. Not the gap in the literature: the thing in the world that the gap is about.
- The question the study asked, stated plainly.
- What was done: who or what was studied, how many, over what period. One or two sentences, with the single design detail that makes the result credible.
- What was found. The main result, with its number and its uncertainty. One result, not five.
- What it changes, and for whom. Keep the claim the size of the evidence.
- Where to read the paper: the title, the journal and the DOI on screen.
Count the words. At roughly 150 words a minute, a two-minute video is about 300 words, which is shorter than most written abstracts. If the draft runs long, cut findings rather than speeding up the narration.
Keeping it accurate
Simplifying is the job, and it is also where a research video goes wrong. The usual failures are specific and avoidable.
- Causal language for a correlational result. If the paper says associated with, the video cannot say causes.
- Dropping the population. A finding in one group, one country or one species is not a general finding.
- Charts redrawn for appearance. Truncated axes, missing units and removed error bars change what a figure says.
- Illustrations that imply a mechanism the study did not test.
- A closing line that promises more than the discussion section does.
The practical safeguard is order of work: every author approves the script word for word before any design begins, and approves the storyboard before any animation. Corrections at the script stage cost minutes. Corrections after animation cost days.
Animation, filmed interview, or screen recording
- Animation suits mechanisms, processes, data and anything that cannot be filmed. It is the most flexible and the easiest to produce in two languages, because the on-screen text and voiceover are replaced without reshooting.
- A filmed interview with the researcher suits work where the person and the setting carry weight: fieldwork, clinical work, community partnerships. It dates faster and is harder to translate.
- A narrated screen recording or slide video is the lowest-cost option and is acceptable to many journals. It is the right choice when the budget is zero and the alternative is no video.
- Mixed: a short filmed introduction with animated figures. Often the strongest format for a lab or project overview.
What funders mean by knowledge mobilization
CIHR's mandate includes mobilizing knowledge into improved health, health services and health systems, and SSHRC publishes guidelines for effective knowledge mobilization for the research it funds. The older term knowledge translation is still widely used in health research and means much the same thing in practice.
Neither agency requires a video. What they ask is that applicants think about who could use the findings and how those people will actually receive them. A video is a reasonable answer when the audience is outside academia, when the result is visual, or when partners need something they can show to others. It is a weak answer when the real audience is twelve specialists who will read the paper anyway.
Whether production costs are an eligible expense depends on the program and on your institution's rules. Ask your research office before you budget for it.
Accessibility and language requirements in Ontario
Under the Accessibility for Ontarians with Disabilities Act, the Integrated Accessibility Standards Regulation requires designated public sector organizations and large organizations to make their web content conform to WCAG 2.0 Level AA, with exceptions for live captions and prerecorded audio description. Universities, colleges and hospitals are covered. In practice this means every prerecorded video you publish needs accurate captions.
- Captions: accurate, synchronised and complete. Automatic captions need correcting, particularly for technical terms and names.
- A transcript published with the video, which also makes the content searchable.
- Enough contrast between text and background, and no information carried by colour alone in charts.
- No important text in the bottom strip of the frame, where captions and player controls sit.
- French: federally funded and national projects are often published in both official languages. Plan the French version at the script stage, because French text runs longer and affects timing and layout.
What drives the cost
There is no reliable published price list for research video in Canada, so rather than quote a figure we cannot source, these are the things a quote is built from.
- Length. Cost rises roughly in proportion to finished minutes.
- Custom illustration. Characters, anatomy, maps and equipment drawn from scratch cost more than type, icons and charts.
- Data animation. Figures rebuilt from the underlying data take longer than figures traced from a PDF, and are more accurate.
- Languages. A second language adds translation, a second voiceover and a second set of on-screen text.
- Review rounds. More co-authors means more rounds. Agree who signs off before you start.
- Extra deliverables: social cut-downs, a silent loop, a poster still.
Ask for a written quote that lists the length, the number of review rounds and every file you will receive, so that two quotes can be compared line for line.
Sources
- CIHR: Knowledge Mobilization at CIHR
- SSHRC: Guidelines for effective knowledge mobilization
- Ontario Regulation 191/11: Integrated Accessibility Standards, section 14
- W3C Web Accessibility Initiative: Making Audio and Video Media Accessible
About this guide
Written and published by ShinobiMedia.ca, an Ottawa web design studio. We sell several of the services our guides discuss, so weigh our conclusions with that in mind and check the reasoning rather than the source.
Prices are taken from the published pages they cite and dated. Legal thresholds, platform rules and Google's guidance are checked against the primary sources listed above. Drafts are prepared with the help of AI writing tools, and every figure, threshold and date is checked by a person before it is published.
If you find something wrong or out of date, email [email protected] and we will correct it and update the date at the top of the page.
Related reading
- Our research video and video abstract service
- Motion graphics and animated explainer videos
- AODA website compliance in Ontario
- Ottawa web design pricing, published in full
- Ottawa web design by ShinobiMedia.ca
Frequently asked questions
How long should a video abstract be?
One to three minutes, and most work best near two. That is a script of about 300 words at a normal speaking pace. Check the journal's author guidelines first, because some publishers set a maximum length and file size.
Do journals require a video abstract?
Very few require one. Many accept them, and some host them on the article page. Whether your journal does, and in what format, is set out in its author guidelines.
Can I make a video abstract myself?
Yes. A narrated slide recording made with the software you already have is acceptable to many journals and far better than no video. Professional production is worth it when the audience is outside your field, when the figures need redrawing to be understood, or when the video will be used by a funder or a partner organization.
What is the difference between knowledge translation and knowledge mobilization?
In practice, very little. Both mean getting research findings to the people who can use them, in a form they can use. Knowledge translation is the older term in health research, and knowledge mobilization is the term CIHR and SSHRC now use.
Does a research video need captions?
If it is published by an Ontario university, college, hospital or other organization covered by the AODA web requirements, yes: prerecorded video has to be captioned to meet WCAG 2.0 Level AA. Captions are good practice regardless, since a large share of social video is watched without sound.
Who owns the finished video?
That is set by your contract with the producer, so read it. You should receive the final files with the right to publish them anywhere, including on a journal's site, and you should confirm whether the journal asks for a licence. Music and stock elements need licences that cover that use.