An Introduction to Cookies and Tracking Pixels

Interpreting your data becomes easier and more meaningful if you understand the pieces that make data collection possible. Every action you take on the web is tracked one way or another. The pages you view, the files you download, and even demographic and interest data can be recorded, and this data is captured through the use of what we call “cookies” and “pixels”. A cookie is a small file that a website stores on your computer. This file might contain the pages you visited and when, a unique identifier, and even if you’re authorized to view certain logged in content. Typically a cookie doesn’t have much-identifying information. The website itself will store its own corresponding file and match your cookie ID to the records on the server, and this is useful for knowing information like: Is the visitor returning for the second time? How long between his visits? And even what advertisements they’ve clicked on.

Speaking of advertising, Ad platforms such as Facebook Atlas or Google AdSense use cookies to identify the same user as they browse the web. So you might see an Ad from Google on the New York Times, and then another Ad from them while you’re browsing a food blog. As you continue to browse the web and load advertisements from the same publisher, they’ll see a list of the type of sites you visit and how you interact with these ads. This information can then be used to sell ads that you’re likely to engage with. So beyond cookies, we have the Tracking Pixel which stores information on your web server, not your computer. A Pixel is really just a one by one transparent image, and they’re often used to see if users convert after visiting a particular Ad.

The server stores a small file called pixel.gif, and every time the server asks for the file, it’s going to log that request. Now instead of just asking for pixel.gif, we will instead add custom variables to the requests such as “pixel.gif? ID=123&ORD ER=ABC”.

The server will be successful in delivering the image because all that extra text is irrelevant, so the user won’t see any issues. However, it does keep a log of that unique URL, and it can use those variables to match that transaction back to a particular user or advertising event. To put this another way, let’s say you click on an Ad to buy movie tickets. As soon as you click, you’re going to receive a cookie. The cookie will include information on the time where you clicked, what banner was clicked, and so on. At this point, the advertising platform knows that the Ad received an impression and a click, but it has no record of the sale. Now let’s say you continue on and buy those tickets. On the confirmation page, the website is going to read the cookie on your computer, pull the information from it, and send it back to the server through the conversion Pixel, and now the advertising platform will connect the dots and indicate a sale for that ad.

We will see more details about how these sales are attributed in later articles. There are some other factors such as how long it has been since the click. If the latter came directly from the last click or maybe if a user saw an advertisement and purchased without actually clicking. We could go deeper and deeper into the technical aspects of how cookies and pixels work, but the truth of the matter is that you really don’t need to know all the nuances. With this high-level overview, you’ll be able to make sure yours are set up properly and tracking the right information.

Leave a Reply