
SAWDUST / BLOG
Choosing music for a data platform explainer video
Music in a data platform explainer video has one job: help the viewer follow how a system they cannot see actually works. Keep it well under the voiceover, change it only where the story changes, and make sure the film still works with the sound off, because on some feeds it starts that way.
Should a data platform explainer video have music at all?
Usually yes, on a short lead. In a study by Moreno and Mayer in the Journal of Educational Psychology (2000, Vol. 92, No. 1), college students watched narrated animations explaining how lightning forms and how hydraulic brakes work, some with background music added. In the lightning study, the music was "an instrumental music loop" that "was synthesized and bland." Students who heard it "remembered significantly less verbal material when music had been presented", and on the problem-solving test the authors found that "Adding extraneous auditory material in the form of music tended to hurt students' understanding of the lightning process."
That was a classroom study, not an ad, and the authors describe the music as bland. But a data platform explainer asks its viewer to do the same work: hold a sequence in their head while a voice explains it. The lesson we take is not "no music". It is that music which is only there to fill silence costs attention. Every bar should do something for the explanation: set the pace, mark a new stage, or give a result room to land.
How loud should background music be under a voiceover?
Lower than most editors first set it. The clearest published figure we know comes from accessibility guidance. The W3C's Web Content Accessibility Guidelines, success criterion 1.4.7 (Level AAA), covers prerecorded audio-only content that "contains primarily speech in the foreground". One way to meet it is for background sounds to be "at least 20 decibels lower than the foreground speech content", with the exception of "occasional sounds that last for only one or two seconds." The same page says background at that level "will be approximately four times quieter than the foreground speech content", and gives the reason: "People who are hard of hearing often have great difficulty separating speech from background sound."
The criterion is written for audio-only content, not video, and Level AAA is the strictest tier. We use it as a mixing reference, not a rule. Under dense technical narration, keep the music far below the voice and let it rise only in the gaps between sentences. In our view, texture matters as much as level: a part that is busy in the same range as the narrator's voice is harder to sit under speech than one that leaves that range open.
How do you give an invisible system a shape?
Write the film's sequence in plain words before anyone listens to tracks: where the data comes from, what happens to it, what the customer sees at the end. Then let the music change at those stage changes, and nowhere else.
A repeating pulse can carry the viewer from incoming data through a system diagram. A new texture can arrive as each new source joins. When the narrator explains a step, the arrangement should hold still. If the music shifts while the narrator is halfway through a point, the viewer hears a new idea that the picture does not support. When the result appears, the voice pauses and the music opens up for a bar or two before the next line.
What happens when the video plays with the sound off?
On some platforms, video starts silent by default. As of November 2026, LinkedIn's video ads page says "LinkedIn Video Ads will autoplay in the feed with the sound muted by default" and that "captions are strongly recommended for LinkedIn Video Ads."
So if the film will run as a LinkedIn ad, the opening has to work twice: as picture and captions for the muted viewer, and as sound for the viewer who turns it on. Do not let the music carry information the captions cannot. A rising build that says "the problem is solved" means nothing to a muted viewer unless the picture says it too.
Example: a 90-second data platform explainer
This is a hypothetical example. A data platform explains how it pulls sales, product and support data into one view.
| Time | What the picture shows | What the music does |
|---|---|---|
| 0:00–0:08 | A dashboard with numbers that disagree | A single low pulse, no melody |
| 0:08–0:30 | Sources flow into the platform diagram | One new texture enters as the data starts to flow, under the voice |
| 0:30–0:55 | Processing, explained step by step | Holds still: no new parts while the narrator explains |
| 0:55–1:05 | The result: one clean view | Voice pauses; the music opens up for four bars, then settles as the voice returns |
| 1:05–1:25 | A customer team using the interface | Back down under the voice, pulse thinned |
| 1:25–1:30 | Logo and call to action | A hard ending on the cut |
The music changes at five section boundaries in 90 seconds, and each change lines up with a change in what the viewer needs to understand.
How is this different from a product launch film?
A launch film sells a moment: the reveal, the name, the date. Our piece on choosing music for a tech product launch video covers that. An explainer has a different viewer, someone trying to understand how the product works, and that changes the music's job from building excitement to keeping the explanation clear. A security film has its own pressure towards alarm, which we cover in choosing music for a cybersecurity product launch video.
Checklist before you lock the music
- Write the system's sequence as one plain sentence per stage.
- Mark music changes only where the stages change.
- Mix the music well under the voice, then check it on laptop and phone speakers.
- Watch the whole film muted with captions on. It should still make sense.
- If the track loops a short section, check whether the repeat is noticeable by the end.
- Brief the music in words: tempo range, where it lifts, where it must stay out of the way. Our guide to writing a music brief for an ad or brand film has a template.
- List every place the film will run, from the website to sales decks and event screens, and confirm the licence covers each one.
Sawdust is a music licensing partner for brands, matching underground electronic music to their values, personality and creative direction. If you are scoring a product explainer, tell us about the film.
SAWDUST
BASED IN
Singapore
ASK AI ABOUT SAWDUST
© Sawdust 2026