
Use Figma's image-resampling option when the exported image needs to be recalculated for a different size. Judge it using the actual output dimensions and the detail a viewer must see. Keep a full-quality source separately; a smaller preview is a delivery rendition, not a replacement master.
Key Takeaways
- Resampling concerns how pixels are recalculated for an output.
- Compare candidates at the intended display size and inspect meaningful details.
- A larger file is not automatically a more useful preview.
Define the preview's job
Write down what the recipient must evaluate. A small gallery preview may need a recognizable product outline and readable key label. A retoucher checking a seam or engraving needs a different file.
For an illustrative woven pouch, identify a few details before comparing exports: the edge of the pouch, the closure, and the label. This avoids judging quality only from an attractive overall impression. The weave might be decorative at thumbnail size but essential in a close material review.
The print pixel-requirement guide explains why output size should come from the destination. Do not assume a website preview and a physical print need the same rendition.
Keep the comparison controlled
Figma's export documentation describes image resampling as useful when exporting imagery at different sizes or converting a vector design to a raster format. For JPG, PNG or PDF export, open Advanced export settings, then the Image resampling dropdown. Detailed, the default, uses bicubic sampling for detail, gradients and shadows. Basic uses nearest-neighbor sampling for hard lines and assets such as pixel art. This is a method choice, not an on/off switch. For the photographic pouch, begin the proposed comparison with Detailed; compare Basic against a named detail requirement rather than assuming a smaller file.
Make clearly named candidates with the intended dimensions. Keep the source composition, format, and color-profile choice fixed while evaluating the resampling change. Otherwise, a difference might come from compression, scale, or color rather than resampling.
Record the settings with the candidate files. A screenshot of a zoomed export preview is not enough to reproduce the result later.
Inspect the saved pixels
Open each candidate independently. Compare them first at the intended display size, then inspect the specific details chosen in the brief. Look for softened lettering, unstable fine patterns, and obvious edge artifacts.
For the pouch example, a tiny preview need not show every thread. It should still represent the shape and important product information truthfully. If a label cannot be read at the required size, redesigning the preview or providing a detail view may serve the user better than repeatedly changing one setting.
Do not expect a small source to regain authentic detail just because it is exported larger. The retained source detail guide explains why preserving the master matters before reduction.
Approve each rendition for its purpose
Check the final file in the page or application where it will appear. Layout scaling can differ from the image viewer. Keep the detailed reference available for users who need it.
If the same image later goes into a PDF, review its effective raster resolution independently. The preview's approval does not transfer automatically to print.
This is a proposed comparison based on the documented export feature. It reports no measured file-size saving or universal best resampling setting.
Related tasks
- Why Does a Photo Stay Blurry After I Shrink It and Enlarge It Again?
- How Large Can You Print a Generated Image?
- Why Is My AI Artwork Still Blurry Inside a Print PDF?
Sources and scope
- Export formats and settings for static designs: Resampling tradeoff for actual output size, distinct from merely selecting a compression format.
Documentation-based guidance. Examples and acceptance checks are proposed, not measured software tests.
Prepared by Imagild Editorial, the company editorial team responsible for these guides. AI assisted research and drafting. This is documentation-based guidance with proposed examples, not a claim of hands-on testing or an independent product review. Report a correction.
Cover: AI-generated conceptual editorial illustration; not a real application screenshot, observed output or before-and-after test.