Stop Exporting One Square Image for Every Platform in 2026
The fastest workflow in social media design is one master file. Export a single 1080×1080 square, upload it everywhere, let each platform do the rest. It is fast, it is tidy, and it is how your headline ends up cropped o
The fastest workflow in social media design is one master file. Export a single 1080×1080 square, upload it everywhere, let each platform do the rest. It is fast, it is tidy, and it is how your headline ends up cropped off on an X header and your logo ends up sliced on a LinkedIn cover.
Here is the argument in one sentence, and it is the whole point of this post: the crop is a decision, and if you do not make it, the platform makes it for you. A 1080×1080 square dropped into a 1500×500 header slot loses its top and bottom. A square forced into a 4:5 portrait feed loses its sides. The platform is not being careless — it is fitting your file into a slot your file was never shaped for. Exporting each placement at its exact size is the only way the crop stays yours.
So which is it: one master file and accept the crop, or exact exports for every placement? I export exact, and below is the table, the source labels, and the ratio math that make it cheap to do.
The 2026 size table I actually use
Every number here comes from the Social Media Image Sizes 2026 guide I built, where each row carries a source label and a last-checked date (September 11, 2026 for official figures). The label matters: "official recommended" and "common creator preset" are not the same kind of number.
| Platform | Placement | Size | Ratio | Source |
|---|---|---|---|---|
| X | Profile photo | 400 × 400 | 1:1 | Official |
| X | Header image | 1500 × 500 | 3:1 | Official |
| Personal cover | 1584 × 396 | 4:1 | Official | |
| Page cover | 1512 × 256 | 5.91:1 | Official | |
| URL preview | 1200 × 627 | 1.91:1 | Official | |
| Standard Pin | 1000 × 1500 | 2:3 | Official | |
| YouTube | Video thumbnail | 3840 × 2160 | 16:9 | Official |
| YouTube | Channel banner | 2560 × 1440 | 16:9 | Official |
| YouTube | Shorts thumbnail | 2160 × 3840 | 9:16 | Official |
| TikTok | Vertical ad asset | 720 × 1280 | 9:16 | Official ad spec |
| Portrait post | 1080 × 1350 | 4:5 | Creator preset | |
| Story / Reel | 1080 × 1920 | 9:16 | Preset + official ratio | |
| Feed / link | 1200 × 630 | 1.91:1 | Creator preset | |
| Story / Reel | 1080 × 1920 | 9:16 | Preset + official ratio | |
| Threads | Portrait post | 1080 × 1350 | 4:5 | Creator preset |
| Bluesky | Landscape post | 1200 × 675 | 16:9 | Creator preset |
Two rows deserve a second look. LinkedIn's URL preview is 1200 × 627 while Facebook's link preset is 1200 × 630 — same 1.91:1 ratio, different pixels, and sharing one file between them means one platform rescales it. And the TikTok rows are ad-network specs from TikTok Ads Manager, not universal rules for organic posts. Using an ad spec for an organic post is not wrong, but it is a choice you should know you are making.
The label is the data
A size chart without source labels is a rumor with a table. Three labels cover almost everything:
- Official recommended — the platform's own Help or developer docs publish that exact size. X's 400 × 400 avatar, LinkedIn's 1584 × 396 personal cover, YouTube's 3840 × 2160 thumbnail.
- Official ad asset spec — the number comes from advertising documentation. TikTok's 720 × 1280 vertical is real and documented — for ads.
- Common creator preset — working creators converge on this export, but the platform never published a simple pixel rule. Instagram's 1080 × 1350 portrait and Facebook's 1200 × 630 link are presets, not hard upload limits.
When a platform publishes no simple current rule — Threads has no first-party pixel table in this review — the honest chart says so instead of inventing a number. That honesty is the feature. A chart that silently promotes a preset to an official limit will eventually tell someone to export the wrong file with total confidence.
Ratio math in a few lines of JavaScript
Twenty rows collapse into four shapes: 1:1, 4:5, 9:16, and 16:9/1.91:1. The ratio is the decision; the pixel count is just the resolution you render it at. A 4:5 export at 1080 × 1350 and one at 800 × 1000 are the same shape at different sharpness. The snippet below is the entire calculation — I use it whenever I need a height for a width I already have:
// Height from a known width and a [w, h] ratio. Rounds to whole pixels,
// because export dialogs do not accept fractions.
function sizeFor(width, [w, h]) {
return { width, height: Math.round((width * h) / w) };
}
sizeFor(1080, [4, 5]); // { width: 1080, height: 1350 } — Instagram portrait
sizeFor(1080, [9, 16]); // { width: 1080, height: 1920 } — story / Reel
sizeFor(1200, [16, 9]); // { width: 1200, height: 675 } — Bluesky landscape
// And the reverse check: does this file actually fit the slot?
function fits(file, [w, h]) {
return Math.abs(file.width / file.height - w / h) < 0.005;
}
fits({ width: 1080, height: 1080 }, [4, 5]); // false — your square is not a portrait
If fits returns false, the platform will crop or pad, and the question is only which one it picks and where it cuts. That is why "close enough" is not a strategy: a 1:1 file in a 4:5 slot is not approximately right, it is a different shape.
The three crops that bite
Most crops are harmless. Three are not:
X header, 1500 × 500. X's own Help article warns that top and bottom cropping can occur. Keep faces and text in the vertical middle of the header, or the crop takes exactly the line you cared about.
YouTube banner safe area. The recommended banner is 2560 × 1440, but the safe area at the minimum 2048 × 1152 size is only 1235 × 338. That central band is what survives across screens. Design the message for the safe area and decorate the rest — a banner designed to the full canvas is a banner designed to be cut.
Pinterest, 1000 × 1500. The standard Pin is 2:3, and Pinterest notes that taller Pins can be cut off in feeds. Taller is not more visible if the feed truncates it. The 2:3 ratio is the ceiling, not the floor.
Fit, Crop, or Pad — pick before you export
Once the size is right, the treatment is the last decision. Fit keeps the entire source inside the target box. Crop to fill makes the exact canvas by cutting edges. Pad to fit keeps the full source and adds space around it. Resizing changes dimensions; compression is a separate step that reduces file weight. If your text sits near an edge, use Pad or Fit — never Crop to fill.
I built ToolkitLoop Image Resizer for exactly this step: open it, type the target dimensions, choose Fit, Crop to fill, or Pad to fit, and download. It runs entirely in your browser — your images never leave your device, no account, no upload. (Disclosure: I built ToolkitLoop, and the size guide and the resizer are my own tools.)
So: one file, or exact exports?
The one-master-file workflow is faster, and for low-stakes posts it is fine. But every time a platform crops your file, it is making an editorial decision about your work — usually across the headline. Exact exports cost a few minutes and keep that decision with you. The table above is the cheap part; the expensive part is pretending the crop will be kind.
What do you do — export every placement exactly, or ship one master file and accept the crop? And if you have been bitten by a specific platform's crop, which one was it?
Originally published by Dev.to WebDev. Aggregated on AIWithGhost for educational purposes — full credit and traffic to the original publisher.