Compress Your Images Without Compromising Quality with Squoosh

Remember when it took forever to download images on dialup? Ain’t nobody got time for that.

Images on the web have been increasing in size throughout the years, and the displays we are viewing these images on have followed suit accordingly. High resolution desktop monitors and large mobile phone displays have become the norm, but unless you’re paying for the fastest internet plan, chances are these large images online will still take a second to load. This is where the recently launched browser utility Squoosh comes in. An open sourced project created by the fine folks over at Google Chrome Labs, Squoosh promises to “make images smaller using best-in-class codecs, right in the browser.”

When you head over to the Squoosh website, you’re given the option to drag & drop one of your images into the window or upload it by clicking on the select an image link. Once your image is uploaded onto Squoosh, the browser based utility will give you the options to resize your image, reduce the color palette, as well as the compression algorithms with which to compress your image. Each compression algorithm has their own specific settings as well, and Squoosh will show you how much space you will be saving when you select each algorithm. You can adjust the quality of the final image before having Squoosh render it for you to download. From our testing, Squoosh worked very well out of the gate and we were happy with the default results the utility gave us. For those interested, Squoosh currently supports the following compression algorithms: OptiPNG, MoxJPEG, WebP, Browser PNG, Browser JPEG, and Browser WebP.

Photographers hosting their own websites will find Squoosh to be especially handy when it comes to fine tuning their Search Engine Optimization (SEO), where the amount of time it takes for your site to load plays a critical factor in how high your site will be listed within search engine queries. The smaller the images, the quicker they load; it’s just that simple.