Your video is ready, but the upload form rejects it. Start with the actual limit and the finished duration. Those two numbers tell you whether a smaller export is realistic or whether you need a shorter edit.
Use the video size budget and acceptance sheet to record the limit, settings, and result. The calculations below are worked examples, not measured compression results.
Start with the delivery requirement
Write down the maximum file size, accepted format, required duration, viewing device, and whether the receiver needs to edit the video. A lightweight review copy and an editing master serve different purposes. If someone needs to recut the original footage, an aggressively compressed copy can create work later.
Check the destination's current requirement directly. A limit on one form does not establish the limit for every email service, messaging app, or social platform. If a review link is acceptable, it may preserve more useful detail than forcing a long film into a tiny attachment.
Save the accepted master separately. Use a name such as product-demo_delivery-v01.mp4 for the experiment. Make subsequent delivery copies from the master, not from yesterday's smaller export. That keeps each adjustment traceable and avoids accumulating another compression step every time you try again.
Choose between a quality target and a size target
HandBrake's quality guide recommends starting with constant quality for ordinary exports and testing small adjustments with a short preview. This is useful when you want a good-looking, reasonably sized file and do not need it to land below an exact ceiling.
A strict size limit is a different job. The constant-quality versus average-bitrate explanation distinguishes a requested quality level from an average data budget. Constant quality produces different sizes for different footage. Average bitrate aims at a data rate, while the resulting visual quality depends on the material.
| Your requirement | Starting approach | What to verify |
|---|---|---|
| Smaller file, flexible final size | Quality-based export, with a short preview | Important detail survives and the size is useful |
| Must fit below a stated limit | Average-bitrate budget with room for audio and overhead | Actual bytes are below the limit |
| Full quality needed for another editor | Keep the master and agree on a transfer method | The recipient has the correct version and format |
Do not interpret a preset name as a guarantee. A screen recording with tiny labels, a handheld shot of leaves, and a mostly still interview stress an encoder differently. Choose the relevant comparison from your own video.
Calculate a useful bitrate budget
For planning, use decimal megabytes and megabits. One byte is eight bits. Approximate total bitrate in megabits per second is the target size in megabytes multiplied by eight, divided by duration in seconds. This includes video and audio; the final container adds overhead.
video Mbps ≈ (target MB × 8 × usable fraction ÷ seconds) − audio MbpsThe usable fraction reserves space. It is your margin, not a codec setting or a guaranteed output size.
Suppose a fictional submission accepts 20 MB and the final clip is 60 seconds. Reserve 10% as a planning margin, leaving 18 MB. The combined stream budget is 18 × 8 ÷ 60 = 2.4 Mbps. If the chosen audio rate is 128 kbps, or 0.128 Mbps, that leaves about 2.272 Mbps for video.
If the useful edit is only 30 seconds, the same 18 MB planning budget permits 4.8 Mbps combined, or about 4.672 Mbps for video with that audio choice. Shortening an unnecessary introduction can therefore create more room per second for the content the recipient actually needs.
| Finished duration | Combined budget | Video allocation |
|---|---|---|
| 30 seconds | 4.800 Mbps | 4.672 Mbps |
| 60 seconds | 2.400 Mbps | 2.272 Mbps |
| 120 seconds | 1.200 Mbps | 1.072 Mbps |
These are arithmetic examples, not recommended rates for every video. If the calculated budget makes faces, motion, or instructions unreadable, change the duration, dimensions, or delivery method. An interface may display MiB instead of MB, and encoders do not necessarily hit an exact byte count. The actual file remains the final check.
Make the export in a controlled order
- Trim first. Remove dead time while preserving the introduction, action, and result the viewer needs. Recalculate using the final duration.
- Choose compatible encoding. Confirm the receiver's supported container and codecs. An MP4 filename alone does not describe the video and audio inside it.
- Set useful dimensions. Test a smaller frame when the intended display does not need the original resolution. Keep the aspect ratio unless reframing is part of the job.
- Choose the rate strategy. Use quality-based encoding for a flexible size or an average-bitrate target for a strict ceiling. Use multipass where the chosen encoder offers it and time permits.
- Check audio and captions. Keep the required track, language, channel layout, and caption information. Do not discard something important to save an unexamined amount of space.
- Export to a new name. Keep the master available until the receiver has accepted the delivery.
HandBrake lists its supported containers, codecs, and preview features. Start with an appropriate preset, then inspect its actual settings. Changing an extension from MOV to MP4 does not re-encode the picture. Likewise, packaging an already compressed video inside a ZIP archive is not a dependable way to achieve a much smaller video.
On an iPhone, Apple documents an iMovie workflow: create a movie from the clip, use the sharing options to export at a lower resolution, and inspect the saved file's information. Keep the original. A convenient mobile export still needs a size and playback check.
Test the most demanding section first
A real export comparison shows why file size is only half the decision. We re-encoded the same eight-second portrait clip on October 22, 2026 with FFmpeg 7.1, the x264 medium preset, and two constant-quality settings. Both kept the original 720 × 1280 dimensions and 24 fps. The AAC audio was copied unchanged. This is one short clip, not a forecast for your footage or a test of HandBrake's interface.
| Copy | Actual bytes | Size | Reduction from source |
|---|---|---|---|
| Existing source | 2,281,398 | 2.281 MB | Reference |
| x264 CRF 26 | 1,074,667 | 1.075 MB | 52.9% |
| x264 CRF 40 | 338,981 | 0.339 MB | 85.1% |
The images below show the same 300 × 300-pixel crop at four seconds, saved losslessly. Open a crop to inspect it at its native size. The CRF 40 version loses more fine cheek texture and shows coarser shading around the nose. CRF 26 preserves more detail, but it is still a lossy re-encode. A smaller number on disk does not establish an acceptable picture.
All three files decoded without errors, and the copied audio payloads matched. That verifies file handling, not subjective quality. A still frame cannot reveal every motion artifact, so play the whole exported copy before accepting it. The measurement notes record the source, settings, method and limitations. These constant-quality exports are separate from the hypothetical 20 MB budget above.
Choose a short section containing the fastest movement and the smallest important detail. For a product demonstration, that might be a hand rotating a labeled bottle. For a tutorial, it may be a menu with small text. A quiet opening frame is a poor test for the rest of a busy video.
Compare that export with the master at the same display size. Inspect text, faces, edges, gradients, and movement. Play it as well as pausing it: a single sharp frame can hide breakup during motion. Listen for missing audio, a changed channel balance, or a timing problem.
A short test helps you judge quality, but its file size does not prove the full export will fit. The full timeline may have different complexity, and quality-based encoding will adapt to it. Export the complete delivery after choosing the settings, then measure that complete file.
Fix the result that actually failed
The file is still too big: confirm you changed the correct export preset and measured the new file. Review duration and total stream allocation. If using a quality target, a smaller size is not guaranteed; switch to a deliberate size budget when that is the real constraint.
The file fits but looks blocky: a larger number of pixels does not compensate for too little information per frame. Try a shorter edit or smaller dimensions, then compare the same difficult section. Do not repeatedly sharpen a damaged export.
The local export looks good but the upload is blurry: move to the source, export, and uploaded-playback quality check. Platform processing and player quality are separate from meeting a file-size limit.
Sound drifts or arrives late: compare the source and export using the audio synchronization diagnostic. A smaller file that loses synchronization has not passed delivery.
How QuestStudio helps
Use Video Lab when the original generated shot needs a clearer subject, simpler motion, or a replacement take. Keep the accepted source and its brief together. Available controls depend on the selected model.
Use your editor or transcoder to make the smaller delivery copy. When the original already works, generating another shot does not solve an oversized export. The AI video quality-control checklist helps separate source defects from delivery defects before you spend time on another generation.
Frequently asked questions
Can I make a video smaller with no quality loss?
Some changes, such as removing unwanted time without re-encoding the retained streams, can preserve the retained material. Lossy re-encoding trades information for size. Judge whether the delivery remains good enough for its purpose instead of assuming it is unchanged.
Does reducing resolution always make the file smaller?
No. Duration and actual bitrate still matter. A smaller frame exported at a high bitrate can remain large. Inspect the complete encoding settings and the saved file.
What bitrate should I use for a 20 MB limit?
It depends on duration, audio, and the space reserved for overhead. The worked 60-second example leaves about 2.272 Mbps for video after a 10% reserve and 128 kbps audio. Treat that as a calculation to test, not a universal quality recommendation.
Should I delete the large original after export?
Keep the master while the project is active and until the delivery is accepted. It is the better starting point for another format, crop, or revision than an already compressed delivery copy.
Finish by reopening the complete file, checking its actual size, and reviewing picture and sound. Record those checks in the delivery budget sheet so the next export starts from a known decision.
Sources and measured export comparison checked October 22, 2026. The 20 MB examples are planning calculations; the eight-second comparison uses actual files. Hero image: AI-generated editorial illustration.