The Three Principles Of Image Optimization

Are you losing visitors to your web site due to slow page load times? Is your site being penalized because the images on your site are too large? Are you able to capture your visitor’s attention in the first 10-15 seconds that they are on your site?

All of these problems can be caused by an improperly optimized web page. In other words, the page loads too slowly and your visitor leaves before you have had a chance to hook them.

Optimizing a page for quick load times can be broken down into 3 broad categories: the basic coding of the page, scripts that are used on the page, and images. Of these three, images that are too large can have the most significant impact on load times and therefore have the greatest potential for improved page loading times if properly optimized.

A brief primer on image types

There are many image formats in common use on web sites, the three most popular being GIF, JPEG, and to a lesser extent Flash content. We will be limiting our discussion to GIF and JPEG images, with a specific focus on still images.

Each of these image formats has their strengths and weaknesses. GIF or Graphics Interchange Format was developed by CompuServe before the Internet boom as a way to share images on the CompuServe service. Due to limitations with screen resolutions and color depths at the time, GIF images were limited to showing up to 256 colors, more colors were imitated by Dithering, a process of fooling the eye into seeing one color by using 2 or more sets of color dots spaced too closely for the eye to distinguish separately.

Imagine a chessboard with black and white squares. When viewed closely we can distinctly see the individual squares, but if we back off far enough we will no longer be able to discern the individual squares and instead we will see one large grey square, the black and white squares merging together in our eyes to form one solid color. This is the concept behind dithering.

The JPEG file format on the other hand is a newer format that can handle millions of colors easily. The initial drawback to JPEG images is that they do contain many more colors, and each color requires some coding for display, making the file size larger.

Speeding up image load times

The main idea behind making an image load faster is to make the file size smaller. This can be accomplished in two ways, you can either make the dimensions of the image smaller, or decrease the amount of coding that is required to display the image.

The easiest way to reduce an images file size is to reduce the image’s physical dimensions. In other words, the smaller the image, the smaller the file size. Imagine an image that is a square 80 pixels by 80 pixels. The number of pixels contained in the image is 80x80 or 6400 individual pixels. If we reduce the image size by one half to 40 pixels by 40 pixels we then have 40x40 or 1600 pixels. So reducing the image size in half reduces the file size to one fourth of the original.

This is our First Principle of Image File Size Reduction: Use the smallest image dimensions that will work with your layout. And likewise the fewer images on the page, the fewer image pixels, therefore the smaller the page size.

Since GIF and JPEG image formats use different methods of saving image information, they tend to be better at showing some types of images and worse at showing others.

GIF images, since they are limited to 256 colors per image, are better at displaying images with large solid blocks of color and images with very small physical dimensions. The GIF format will produce smaller file sizes than JPEG for these types of images.

JPEG images are better at showing gradients or subtle changes from one color to another. Therefore JPEGs reproduce photographs very well, or any other image with gradations. The JPEG format will produce smaller file sizes for these types of images than the GIF format will.

This is our Second Principle of Image File Size Reduction: Choose the correct image format for the image you are using. Most web pages will contain a combination of GIF and JPEG images.

Decreasing the coding is called image compression. Both GIF and JPEG images can be compressed but the process is different. In GIF images we try to limit the number of colors, in a JPEG image we use software algorithms to remove redundant information from the file.

Whenever we compress a file we will lose some image quality. We have to reach a balance between a small file size and acceptable image quality.

This is our Third Principle of Image File Size Reduction: Find the least acceptable level of image quality. Most images can handle some compression with very little quality loss, and all images can stand more image quality loss and still be acceptable. Your job is to decide how much quality loss you can accept. In other words, the lower the quality, the smaller the file size.

GIF images can usually be reduced from 256 colors to 128 colors or less, the fewer colors used the smaller the file size. JPEG images can almost always be reduced to a quality setting of 80% and frequently can be reduced down to as little as 15-30%. So when you use a higher compression level (smaller number) the file size will be reduced. Experiment with the image, try smaller and smaller settings until you find the smallest setting that still displays an acceptable quality.

The fastest loading page will have no images and the slowest loading page will be completely filled with full resolution images. If you work towards controlling your images using the principles outlined above you will have a very lean web page that will load quickly and be viewed favorably by the search engines.