Universal post preview

Start with the part you cannot afford to lose

Choose an image in which the subject, important text, and any required disclosure are easy to identify. Then move the horizontal and vertical focal-point sliders. The eight previews update together, letting you compare a square, a tall portrait, and a wide landscape without repeatedly exporting from an editor. The goal is not to make every rectangle look identical. It is to find where the original composition stops working and decide whether that destination needs a separate design.

The input stays in this browser. Frenzlab does not upload the image, send it to an image-processing service, connect to an account, or ask a platform what it would publish. A JPEG, PNG, WebP, or AVIF is accepted up to the tool's own 15 MB and 40-megapixel limits. These are local safeguards, not social-platform upload limits. If decoding fails, export a smaller JPEG from your editor and try again.

What the crop calculation actually does

A crop is a rectangle taken from the source image. The tool compares the source aspect ratio with the chosen output ratio. If the source is wider, it removes width; if the source is taller, it removes height. It never stretches the subject to make the numbers fit. Moving a focal point changes which portion is retained, not the dimensions of the retained rectangle.

For a target width divided by height of R, the retained source width is the smaller of the original width and the original height multiplied by R. The retained height follows the same calculation in the other direction. The excluded percentage is one minus retained area divided by original area, multiplied by 100. That percentage describes area, not the importance of whatever was removed.

The output canvas then resizes the retained rectangle to the template dimensions. The downloadable JPEG includes the image alone, not the planning overlays. Transparent pixels are placed against white during export. Keep the original if you need a transparent deliverable; this tool's JPEG output is not a substitute for it.

Read templates as working canvases

The format names are intended publishing contexts. They are not a claim that every current app screen uses these precise proportions or safe margins. Check the destination surface in the publishing app. Instagram feed, profile-grid, and Reels presentation can differ. [VERIFY: Instagram Help Center — image resolution and profile-grid presentation — https://help.instagram.com/]

The tall-template overlays use a fixed ten-percent side margin and an adjustable top and bottom margin. They are editorial planning guides, not measured platform controls. A smaller safe area does not make your image safer automatically; it simply asks you to keep essential material in a tighter region. Test labels and subtitles against the actual interface before publishing.

Three worked examples

Example 1: A square becomes a tall frame

Start with a 2000 × 2000 source and choose a 9:16 output. The retained width is 2000 multiplied by 9 divided by 16, or 1125 pixels. Height remains 2000. You keep 2,250,000 of 4,000,000 source pixels by area, so 43.75 percent is excluded. At the midpoint, 437.5 source pixels are removed from each side. Moving the horizontal slider shifts that retained column; it does not restore the missing width.

Example 2: A landscape photograph becomes a feed portrait

A 3000 × 2000 photograph cropped to 4:5 retains 1600 × 2000 pixels. That is 3,200,000 of 6,000,000 pixels, leaving 46.67 percent excluded. If a product occupies the right third, move the horizontal focal point toward it and inspect the wide template too. You may find that the portrait looks better while the square loses a supporting object. That is a design decision to resolve, not a software error.

Example 3: A portrait becomes a square

A 1200 × 1800 source cropped to 1:1 retains 1200 × 1200 pixels. One third of the area disappears. Centering removes 300 pixels from the top and 300 from the bottom. If a headline is near the top, move the vertical focal point upward. If that hides the product at the bottom, the source does not contain enough room for both at this ratio. Make an alternate layout instead of shrinking essential text until it is unreadable.

Inspect the source, not only the result

The source inspection control shows a white rectangle over the chosen format's retained area. Darkened regions are excluded. Switch this control between a narrow and a wide output to see why one shared focal point is sometimes insufficient. The image dimensions printed beneath the controls refer to the decoded source, while each preview's dimensions refer to its download canvas.

Try the built-in geometric sample before using a client asset. Its large text and contrasting shapes make losses obvious. After experimenting, choose your actual file, check all eight frames, and download only the formats you need. A download does not publish anything or create a scheduled post.

The most common mistake: trusting a centered crop

The midpoint is a convenient starting position, not an editorial recommendation. A centered rectangle can preserve a background while cutting off the thing the image is supposed to communicate. Choose the subject first, then inspect text, hands, packaging, and any meaningful edges. Area retained is a useful measurement, but a crop keeping ninety percent can still lose the crucial ten percent.

What this tool deliberately does not do

There is no platform login, video processing, face detection, generative fill, or automatic subject selection. It does not certify that an app will display the exact same crop. That boundary keeps the tool private and its result understandable. Use the export settings advisor to prepare an editor preset and the caption truncation previewer to rehearse accompanying text. For the broader workflow, read the image and video sizing reference.