Start with the upload cap, not the file format
5 MB is the limit that breaks a lot of iPhone photo uploads. A 12 MP HEIC from an iPhone 15 can be only 2 MB to 4 MB on the phone, then become a 5 MB to 8 MB JPG after conversion because JPEG is less efficient. That does not mean the conversion failed. It means the delivery copy needs a size target.
Start with the main HEIC to JPG converter when the destination only rejects .heic. If the JPG is still too large, use image compression before trying another random export. For folder jobs, send the originals through batch conversion, then optimize the outputs in smaller groups. Apple has used HEIC as the default iPhone photo format since iOS 11 in 2017, and its HEIF and HEVC support guide explains why the format saves space. The upload site may still want plain JPG.
Pick the right target before converting
A file-size workflow works best when you choose one target and stop editing once the image passes.
2 MB profile for strict forms
Use this for job portals, school forms, ID uploads, and older government systems. Convert to JPG around 82% to 85% quality, then resize to 1600 px or 2000 px wide with the image resizer if the file is still above 2 MB. Strip location and device metadata with EXIF removal before uploading anything public or identity-related.
5 MB to 10 MB profile for normal sharing
Use 88% to 92% JPG quality for marketplace photos, support tickets, real-estate previews, and email attachments. That range usually keeps visible detail while trimming 15% to 35% compared with a 95% export. If you are deciding whether to avoid cloud tools, free alternative to CloudConvert explains the local workflow, and better than Convertio covers repeated upload friction.
How browser compression changes the JPG bytes
Browser-local conversion and compression are separate steps. First, the browser reads the HEIC through the File API, decodes the HEIF container and HEVC image data through native support or a WebAssembly codec, then draws the pixels into a Canvas-style buffer. The W3C File API describes the browser file access layer; the High Efficiency Image File Format background explains the container side.
After decoding, JPG export applies lossy compression. Quality settings change how much image detail is approximated, while resizing reduces the pixel count before export. A 4032 x 3024 image has about 12.2 million pixels; resizing it to 2000 px wide cuts that to roughly 3 million pixels. That single resize can do more for a 2 MB cap than lowering quality from 90% to 70%.
A clean workflow for stubborn uploads
Use one pass for each type of change. Convert the HEIC, check the JPG size, then resize or compress only if the destination rejects it. The detailed JPG quality guide is useful when the file looks fine but lands too large.
- Keep the original HEIC as the archive copy.
- Convert through / or the batch converter.
- Try
88%quality for general uploads or84%for a 2 MB cap. - Resize once if the dimensions are excessive.
- Remove EXIF when the image will leave a trusted workflow.
This order avoids repeated lossy exports. It also keeps the privacy decision separate from the size decision, which matters for photos with GPS, timestamps, or device model data.
FAQ
Why is my JPG bigger than the HEIC original?
HEIC usually stores similar quality in 30% to 50% less space than JPG, so a 3 MB HEIC becoming a 6 MB JPG is normal. The fix is targeted compression or resizing, not repeated conversion.
What JPG quality should I use for a 5 MB limit?
Start around 88%. If the image is still over 5 MB, resize the long edge to 2000 px before dropping quality below 82%, because pixel count usually drives size more cleanly than harsh compression.
Does resizing hurt photo quality more than compression?
Resizing changes dimensions, while compression changes detail inside the existing dimensions. For upload forms that display images at 1200 px or 1600 px, resizing from 4032 px wide is usually the cleaner trade.
Should I remove EXIF before or after compression?
Remove EXIF after you have the final delivery file. Metadata cleanup is separate from JPG quality, and one final pass is enough for most public uploads.