Video content
Effective video combines clear sound, purposeful visuals and a story suited to its audience. A writer helps shape the content, prepare interviews and scripts, and coordinate the words, images and timing of the finished piece.
Purpose and production planning
A video should serve its content: if suitable visuals add nothing to the story, another medium should be used. Producing a science video requires four things:
- A thorough understanding of videography.
- Videography equipment.
- An even better understanding of the paper's subject.
- A creative idea for conveying the message.
Medical writers bring subject understanding and the ability to communicate key messages, while videography requires additional learning. Watching effective examples and attempting similar techniques can help develop production skills; one beginner spent two to three hours weekly doing this. Suggested examples included:
- TED-Ed.
- AsapSCIENCE.
- minutephysics.
- NIH videos.
- PBS Nova Profiles.
A design report provides a preproduction plan, including an outline that divides the information into manageable pieces and sequences them, for example from simple to complex or general to specific. With the outline established, the team selects the medium, production style, content, organisation, tone and language level for the audience and message. The creative concept defines the storyline, using established approaches where appropriate, such as an illustrated lecture combining a presenter with pictures.
A suggested length for a short science video is two to three minutes, or four to five minutes when the material is particularly engaging or important. There is no fixed word requirement, though one estimate allowed about 600 words for a three- to four-minute video.
Story and audience engagement
Three principles guide one approach to video, with story given the greatest emphasis:
- Good sound.
- An effective on-screen personality.
- A story.
Novelty alone can lose its appeal: the Lumière brothers' raw-footage films attracted paying audiences in 1895, while by about 1900 most of their shorts had a story. A comparable change was described in 2012 corporate 3D production, where clients increasingly wanted narration and story rather than footage demonstrating the technology.
A story establishes an expectation, makes the audience wait and resolves it at the end. It can sustain interest beyond the initial attraction of the audio and presenter, and can work outside fictional cinema or television drama.
A hypothetical asthma-drug promotion begins with a child unable to join friends at play, introduces the company and treatment, and ends with the child playing with the others, presenting a benefit rather than opening with product features. The narrative structure establishes a character and a situation the audience wants resolved.
A hospital-equipment review can instead create suspense with a question: how good is a new MRI scanner? Following the presenter to a hospital for a scan makes the audience wait for the answer through an on-location report. Blendtec's “Will it Blend?” videos likewise established an expectation in their title and resolved it by blending a different object each week.
The storyline can be adapted to the audience: sales-representative material might move from disease and existing treatment to the new medicine and its differences, while patient material might move from disease to outcomes and choosing a treatment option.
Sound and the presenter
Poor sound can make viewers stop watching even when the picture is acceptable, while clear audio can retain them despite imperfect visuals. The microphone should be close to the sound source. For a busy-street recording, a lapel microphone stays close to the mouth, while a directional microphone should point towards it, typically from below, avoiding other sound sources behind it. A laptop's built-in microphone may suffice for a webcam video, but the camera's internal microphone is discouraged where it is too distant and captures unwanted sound.
Checking that the microphone works is essential: poor video can be replaced with a still photograph or B-roll, but missing audio cannot be recovered that way.
An on-screen presenter can help connect with the audience. Enthusiasm, gestures and a voice louder than ordinary conversation can convey energy and expertise, balanced with a genuine manner. Screen presence takes practice and honest feedback, and another presenter can be used if they communicate the message better. Speaking with authority, smiling and remaining positive were common features of two effective examples with very different production budgets.
For short science videos, another recommendation is to avoid talking-head footage or limit it to about 30 seconds against a clean, tidy background.
Interviews and editing
Interviews supported by B-roll can engage viewers, but effective interviews require substantial preparation. Focused questions, rather than broad open-ended ones, help obtain answers that follow the planned storyline.
Interviewees should use complete sentences, incorporate the question into the answer to provide context and keep responses brief. Strong opening and closing quotations are useful in documentary-style work. The writer selects sound bites and assembles the story from interview passages and B-roll: interspersed pictures that add interest, background and context.
Screen text, visuals and pacing
On-screen sentences need to be short because viewers may have little time to read them. A recommendation for two- and three-dimensional animation is to use voiceover rather than rely on subtitles to guide the narrative and draw attention to important features.
A still image can remain visible for ten to fifteen seconds or longer when accompanied by speech, though about seven seconds was one suggested average. The “Ken Burns effect” focuses on part of an image and then zooms out. One production recommendation advises against light text on a dark background.
Free background music can be found online; Longzijun was a recommended source in 2017. An asset-search example used “lungs.obj free” to locate object files for illustration or animation.
A branding recommendation includes a watermark and five-second branded sections at the beginning and end.
An educational animation workflow
A 2024 PHUSE project developing public-facing clinical-trial data-literacy videos used five steps:
- Scriptwriting.
- Audio considerations.
- Illustration and storyboard creation.
- Video animation and audiovisual integration.
- Dissemination.
The scripts were developed with patient-advocate feedback, using first-person dialogue, contractions, concise sentences and explanations of technical terms to create an approachable conversation. The team consistently used “trial” rather than “study” and repeated “safe” and “effective” as key terms. An example opens with one fictional character asking another why she is interested in a trial, and the reply that she has a condition and wants other treatment options. Drafts were reviewed by the project team and PHUSE leadership, amended and proofread before production.
For audio, the team reached consensus on a human voice actor through a survey and reviewed and approved the recording. AI voiceovers were also tested to reduce the cost of lengthy, complex videos and explore choices of voice, character and translation beyond English. Repeated revisions still left problems with robotic or impersonal delivery, insufficient pauses, excessive speed and unusual intonation or inflection. After consulting patient groups, the team returned to human voices, judging the tested AI output in 2024 inadequate for the desired empathy, tone and contextual understanding.
Character expressions were added to reflect varied emotions and address concerns that digital materials often miss patients' complex feelings about trials. Animation began with character movements, backgrounds and integrated narration, followed by review of where additional pauses were needed and precise timings for the videographer. Facial movements were added with patient-advocate feedback before finalisation, and a running transcript accompanied the dialogue with keywords appearing in prompt boxes.
Viewers could pause and repeat complex sections; varied images, highlighted words and short pauses between concepts were used to help them process the words, sound and visuals.
Recording a live presentation
A 2007 webcast guide outlined four production considerations:
- Equipment: a standard camcorder could replace a professional camera, with an external microphone for reliable audio; a video encoder linked it to the computer, and Windows Media Encoder, then available free, supplied the encoding software.
- Recording: the speaker was asked to minimise movement, and the camera remained on a tripod; experienced operators could use close-ups and varied angles.
- Preparation for posting: longer presentations produced larger files, which could be transferred by FTP or flash drive; the encoding software supported editing, an audiovisual switcher could insert clips during encoding, and the output played in Windows Media Player.
- Further development: video could run alongside synchronised, automatically advancing PowerPoint slides, potentially requiring a video specialist and webmaster.