
I found myself in a debate with a colleague the other day regarding file sizes for advertisement banners on a website. We are in the process of setting image size specifications for static advertisement banners to give to ad vendors who will be creating the final ads. Based on the width and height of the ads in question I proposed a conservative yet realistic file size of 20kb, just to keep sizes down where possible. I was met with some resistance, and an argument that went something like “with all the broadband connections out there, why would we not allow for 50kb so that we can get the best image quality?”. My view on this is actually the other way around. Why would you not optimize your website wherever possible? Why not force the smaller file sizes (which were well within reason to begin with) so that page load, file storage, bandwidth, etc. can all benefit, even if just a tiny bit, from the savings. It all adds up.
Let me use an example outside of the web to draw a comparison. When Mazda was redesigning their latest model of the Miata MX5 a few years ago, they were faced with a car that was going to be larger, with more safety features, and would represent a vast improvement from the previous model. However, with a small, great handling car like the MX5, weight is the enemy – and this car could have easily gained hundreds of pounds through the model change. Mazda tasked all of their engineers and designers to shave weight wherever they could. They saved a few grams off of redesigning the rear view mirror, for example. They did this with hundreds of components within the car. In the end the new model only ended up weighing 27 pounds more than the previous model. An amazing accomplishment considering what was involved to get it there.
So back to my point. 20 or 30kb here or there may not seem like much. But when you account for thousands of page loads and potentially hundreds of banners, it eventually all ads up (pun intended). We should always adhere to best practices, even if technology seems to grant us more leeway than we are used to having.
Written by Matt Glaze