Exporting Photos: Formats, Color Space, Size and Sharpening for Every Destination

Exporting turns your edited master into a finished file built for one destination: a website, a phone screen, a print lab or a client. Every export answers the same five questions: which file format, which color space, how many pixels, how much compression and sharpening, and which metadata travels with the file. Get those right for the destination and the photo looks the way you edited it; get them wrong and the same image can arrive dull, soft, blocky or carrying your home address.

Why the master and the export are different files

Your working file is built for editing. It might be a raw file with a list of instructions attached, or a layered file with adjustment layers and masks. It holds more color, more tonal precision and more pixels than any single destination needs, and it can be changed at any time without loss. That is the whole point of non-destructive editing.

An export is the opposite: a flattened, finished copy with every decision baked in. It is made smaller, converted to the colors the destination understands, sharpened for the way it will be seen, and stripped of anything the viewer does not need. Because it is disposable, you can make as many as you like from one master: a small one for a website, a large one for a print, a medium one for a client proof.

Two habits follow from this. First, never edit an export and then export it again. Each lossy save throws away a little more information, and repeated saves compound the damage. Go back to the master instead. Second, keep exports out of your master folders, or name them so clearly that nobody (including you, years from now) confuses a 2,000 pixel web copy with the original.

Choosing a file format for the destination

The format decides how the pixels are stored, and each one makes a different trade between file size, quality and compatibility. The image file formats guide covers the internals; for exporting, what matters is which one to reach for.

• JPEG is the default for sharing. Every browser, phone, operating system and print kiosk opens it. It uses lossy compression, which discards detail your eye is unlikely to miss, and it stores only 8 bits per channel. For a finished photo that will not be edited again, that is almost always enough.
• TIFF is the safe choice when quality matters more than size: fine-art print labs, publishers, and archival copies of a finished edit. It can be uncompressed or use lossless compression, and it can hold 16 bits per channel. Files are large, often five to ten times the size of a good JPEG.
• PNG is lossless and supports transparency. It is ideal for screenshots, logos, graphics and text, but a full-resolution photo saved as PNG is huge for no visible benefit. Use it for photos only when a destination demands lossless files and does not accept TIFF.
• HEIF and AVIF are more efficient formats that deliver similar visual quality to JPEG at noticeably smaller sizes, and they can store more than 8 bits per channel and high dynamic range. Their weakness is compatibility: not every browser, program or print service opens them, so keep a JPEG version for everything else. See HEIF and HEIC for more.

If you do not know what the recipient will open the file with, send a JPEG.

Color space: sRGB unless you have a reason

A color space defines the range of colors a file can describe and what its numbers mean. The same pixel value of, say, 200 red means a more intense red in a wide space than in a narrow one. That is why the color space has to be converted, not just labeled, when you export, and why the file must carry an embedded profile that tells the viewer which space it is in. The color space explainer goes into the details; the export decision is simpler.

Use sRGB for the web, social platforms, email, messaging, slideshows, most consumer print labs and anything whose destination you do not know. sRGB is the space the web assumes and that nearly every screen can show. A file in a wider space viewed by software that ignores profiles will look flat and desaturated, because the numbers get read as if they were sRGB.

Use a wider space (Adobe RGB, Display P3 or ProPhoto RGB are the common ones) only when the destination is color managed and can use the extra range: a fine-art lab that asks for it, your own inkjet printing from a color-managed workflow, or a client who will edit further. Wide spaces pair with 16-bit files. Spreading 8 bits of tonal steps across a very wide space leaves bigger gaps between them, which shows up as banding in smooth gradients after further editing.

When your master contains colors outside sRGB, typically saturated reds, oranges, cyans and deep greens, converting to sRGB has to squeeze them in. Most editors can show you which colors will change before you commit, the same idea as soft proofing for print. If a sunset turns into a flat orange blob on export, pull saturation back on those hues in the master rather than accepting the clipping.

Pixels, ppi and resizing

Screens care about pixel dimensions and nothing else. A photo 2,000 pixels wide fills about the same space on a given screen whether its file says 72 ppi or 300 ppi. That ppi number is only an instruction for printing. The difference between ppi and printer dots is covered in DPI vs PPI.

For screens, size by the long edge in pixels. A long edge of about 2,000 to 2,500 pixels looks crisp on most large displays without making the file heavy. Social platforms resize and recompress everything you upload, so matching their recommended dimensions (listed in social media photo sizes) gives their software less to mangle.

For print, work backwards from the print size. Pixels needed equals inches multiplied by ppi. An 8 by 10 inch print at 300 ppi needs 2,400 by 3,000 pixels; the same print at 240 ppi needs 1,920 by 2,400. Around 300 ppi is the common lab request for prints viewed at arm’s length. Large prints are viewed from farther away, so lower values work there, which is why a poster does not need a gigantic file. The print size calculator does the arithmetic and image resolution and file sizes explains the relationship in depth.

Resizing means resampling: the software calculates new pixels from the old ones. Two principles cover almost every case:

• Making an image smaller is safe and usually improves apparent sharpness per pixel. Methods tuned for reduction keep edges crisp.
• Making an image larger cannot create detail that was never recorded. Methods tuned for enlargement produce smoother results, and modest enlargement for print is fine, but doubling the size of a small file will look soft up close.

Compression quality and file size

JPEG and the efficient formats let you pick a quality level. Higher quality keeps more fine detail and smoother gradients; lower quality makes smaller files. The scale is not standardized, so 80 in one program does not produce the same file as 80 in another. What holds across editors is the shape of the trade:

• From maximum down to roughly the upper third of the scale, file size drops sharply while visible quality barely changes. This is the zone to live in for most sharing.
• Below the middle of the scale, artifacts start to appear: blocky patches in smooth skies, smeared fine texture, and faint halos or “mosquito noise” around sharp edges.
• Maximum quality gives the largest file for a difference few people can see. Use it for files that will be printed or handed to someone who might edit them, not for the web.

Content changes the right setting. Busy scenes full of leaves, rock or crowds hide compression well. Smooth gradients such as clear skies, fog, studio backdrops and night skies reveal it first. The practical test is to export the same photo at three quality levels, open them at 100 percent side by side, and pick the lowest one you cannot tell apart from the highest. Save that as your default, and bump it up for images with large smooth areas. The mechanics are explained in how JPEG compression works.

Output sharpening for screen and for print

Resizing softens an image a little, and every medium softens it again in its own way. Output sharpening is the last step, applied at the final size for the final medium, and it is separate from the capture and creative sharpening you did while editing. The full reasoning is in the sharpening guide; here is how it applies at export.

For screens, use a light amount. A downsized image already looks crisp, and screens show every pixel exactly. Too much produces bright halos and a crunchy texture. Judge it by viewing the exported file at 100 percent on a normal screen.

For print, use noticeably more. Ink spreads slightly into the paper, and the print is viewed from a distance where fine edge contrast gets lost. Matte and textured papers soften more than glossy ones, so they need more sharpening. The catch is that correct print sharpening looks too strong on screen. Do not judge it at 100 percent on your monitor; judge it on a test print at the real size, then adjust your preset.

If you export for a lab that applies its own sharpening, ask whether to send files unsharpened. Sharpening twice is a common cause of harsh prints.

Metadata to keep and metadata to strip

Every export can carry information alongside the pixels. It falls into three groups, explained further in editing photo metadata:

• Camera data (EXIF): camera, lens, shutter speed, aperture, ISO, date and time. Harmless for most photos and useful for people who want to learn from them.
• Descriptive and rights data: your name, copyright notice, contact details, caption and keywords. Worth keeping on almost everything, because it travels with the file when it is downloaded or forwarded.
• Location data: GPS coordinates recorded by the camera or phone. Precise enough to identify a house, a school or a regular walking route.

A sensible default is to keep your name, copyright and caption, and to remove location from anything shot at or near home, of children, of wildlife that should not be disturbed, or of anyone who has not agreed to share where they were. Sharing photos online safely covers the privacy side in depth.

Note that many social platforms strip most metadata on upload, so embedded copyright is not reliable protection there, and a visible watermark or your own site may serve you better. Removing metadata from an export leaves the master’s full record intact.

Naming and organizing exported files

A file called IMG_4821-2-edit-final.jpg tells nobody anything. A naming pattern that sorts correctly and describes the file saves time for you and whoever receives it. A reliable pattern is date first, then subject, then a sequence number or size tag: a four-digit year, two-digit month and two-digit day, followed by something like _lake-sunrise_web.jpg. Dates written year, month, day sort in order in any folder view.

• Use lowercase letters, numbers and hyphens or underscores. Spaces and special characters break some web links and servers.
• Add a suffix for the destination (web, print, client) when you keep several exports of the same photo.
• Export into a dedicated folder per destination or project, so a whole set can be uploaded, zipped or deleted at once.
• Keep the original file number in the export name, so you can find the master from any copy.

For a complete folder and backup system, see photo file management.

Export settings by destination

Almost every editor lets you save export settings as presets. Build one per destination from the table below, test each once, and you will never again send a web-sized file to a lab or a 60 megabyte file by email.

For a deeper look at the web side, including how browsers and platforms treat your files, see how to make your photos look great online. For the print side, see preparing photos for print.

Common mistakes

• Exporting web images in a wide color space. They look dull in any viewer that ignores profiles. Fix: convert to sRGB for every screen destination and embed the profile.
• Editing and re-saving the JPEG export. Each save loses more. Fix: make changes in the master and export a fresh copy.
• Uploading full-resolution files to the web. Pages load slowly and the platform recompresses them harder. Fix: resize to the long edge the destination needs.
• Changing ppi and expecting a sharper screen image. Screens ignore it. Fix: change pixel dimensions for screens; use ppi only to calculate print sizes.
• Judging print sharpening on the monitor. Correct print sharpening looks overdone on screen. Fix: make a small test print on the real paper and adjust from that.
• Leaving GPS data on photos taken at home. Anyone who downloads the file can read the location. Fix: strip location in your web and social presets by default.
• Squeezing file size with very low quality. Skies turn blocky and gradients band. Fix: find the lowest quality that looks identical at 100 percent and stay at or above it, higher for smooth scenes.

Try this

Pick one finished photo with a large area of smooth sky. Export it four times as sRGB JPEGs at 2,000 pixels on the long edge: once at maximum quality, and then at three lower settings spread down to the middle of your editor’s scale. Note each file size. Open all four at 100 percent and compare the sky and the finest texture. Choose the lowest quality you cannot tell from the maximum, then save a preset called “web” with that quality, light screen sharpening, your copyright and caption kept, and location removed. Finally, check the exported file’s properties to confirm the size, color space and metadata are what you intended.

Frequently asked questions

What resolution should I export for social media, a website or email?

For screens, think in pixels, not ppi. Around 2,000 to 2,500 pixels on the long edge suits websites; email and messaging are fine at 1,600 to 2,000. For social platforms, use their recommended dimensions so they resize your file as little as possible.

Should I export at 72 ppi or 300 ppi?

For screens it makes no difference; only the pixel dimensions matter. For print, set the pixel dimensions from the print size (inches multiplied by roughly 300) and the ppi value will follow.

Why do my exported photos look washed out online?

Almost always because the file was exported in a wide color space and is being displayed by software that assumes sRGB. Re-export as sRGB with the profile embedded.

Is PNG better than JPEG for photos?

Not for sharing. PNG is lossless, so the file is many times larger with no visible gain for a photo. It is the better choice for graphics, text and anything needing transparency.

Should I send a print lab JPEG or TIFF?

Whatever the lab asks for. Consumer labs usually want sRGB JPEGs at maximum quality. Fine-art labs often prefer 16-bit TIFFs in a wider space. When in doubt, ask them.

Related guides

• Image file formats explained: how JPEG, PNG, TIFF, WebP and HEIF store pixels and when each makes sense.
• Color space explained: what sRGB and the wider spaces really contain and how conversion works.
• Sharpening: capture, creative and output sharpening, and how to avoid halos.
• How to make your photos look great online: what happens to your files after they leave your computer.
• Editing photo metadata: what EXIF, IPTC and XMP hold and how to change them.
• Sharing photos online safely: location data, faces and what not to post.
• Preparing photos for print: sizing, proofing and paper choices before the lab.
• Photo file management: folders, naming and backups for masters and exports.