Expires headers tell the browser whether they should request a specific file from the server or whether they should grab it from the browser’s cache.
The whole idea behind Expires Headers is not only to reduce the load of downloads from the server (constantly downloading the same file when it’s unmodified is wasting precious load time) but rather to reduce the number of HTTP requests for the server.
When you visit a website your browser is responsible for communicating with the web server to download all the required files. It then compiles those files to display the web page. As web pages become richer in graphics and content, more and more files are being transferred between your machine and the web server.
In the past you would have an HTML file and maybe a few images to serve for your website, however many modern websites might have 50+ files per page to transfer. The files themselves can be a huge load increase by themselves but for each file you must create a request and even if requests are fractions of a second, they can soon add up.
How does it work?
Expires Headers are rather simple in how they work. They tell the browser how long to store a file in the cache so subsequent page views and visits they don’t have to download the file again. You are right to assume Expires Headers don’t improve page speed for a first time visit as this visitor would have to download all the files for the first time. Using Expires Headers helps increase load times for returning visitors.
You can set Expires headers on specific files or even file types. Then when the browser comes to the website it can see when was the last time it downloaded the specific file types. If it was recently it will display them from the cache, if you haven’t visited the site in a while it will download the newest version from the web server.
The idea is to set late expiry times for items that don’t change on your website (logo, colours etc). Set short expiry times for things that change regularly.
Why is it important?
Adding Expires Headers is important to reduce HTTP requests which reduces the time it take for the server to communicate with the browser. It also allows your users to reuse the cache files that have been stored in the browser to reduce the amount of files they need to download.