The vocabulary of image optimization.

Image format developed by Google, usually 25 to 35 % lighter than a JPEG with equivalent visual quality. Supported by all modern browsers.

New generation format, even more compact than WebP on most photos. Browser support slightly more recent than WebP, hence the interest of serving an automatic fold.

Google Core Web Vitals Metric that measures the display time of the largest visible item on the screen, often the main image of a product sheet. Lighter images directly improve this score.

Delayed loading: images out of the screen are only downloaded when the visitor approaches them by scrolling the page, which speeds up the initial display.

Each image variant (size, format, mode) is calculated only once and then saved to disk, all subsequent requests receive it instantly, without recalculation.

Ready-to-use configuration (dimensions, resize mode, quality) associated with specific use , product_card, product_page, category_banner for example, to avoid setting each image manually.

Information embedded in an image file (photo camera, orientation, sometimes geolocation). The orientation is useful and corrected automatically; the rest is usually deleted to lighten the file and protect privacy.

Image generation URL accompanied by a cryptographic signature that prevents a third party from requesting arbitrary dimensions or formats not provided for by the shop.

Five ways to adapt an image to target dimensions: Fit and Container preserve the entire image (with or without visible background), Cover and Crop fill the space by recadrant, Stretch distorts the image to the exact dimensions.