As discussed in the last post on the collection of website usage statistics, reviewing website usage statistics is one of the most important tasks for a digital marketer.

It is particularly important to be responsive to usability problems online, where users can easily abandon your site or social media channel if it is not performing well.

Easy access to website traffic and other statistics makes both the web design and marketing processes organic and iterative.

Designs and campaigns can quickly evolve, depending on how effective they are for their end users.

But how is this usage information collected?

Methods of Data Collection

Some of the following methods are used by analytics software in order to determine website traffic and usage:

IP Address

One of the most common clues for discovering more information about the physical location and habits of your website visitors is by recording their IP address.

Issues: IP addresses do not reflect each machine that accesses the site, but rather the public IP address of the network used by each machine. Therefore, you could have several connections from the same IP address that are only registered by your analytics software as one unique visitor.

Cookies

peanut cookiesCookies are small pieces of data stored on a user’s computer by a website. Cookies can be used to assign a unique ID to each user who visits a site, in order to tailor the information presented to that user, or to record their actions.

Issues: Cookies can be denied or deleted by users and according to EU Law and privacy concerns, their use is now subject to gaining the website user’s informed consent.

Page Tags

Also known as tracking bugs, page tags are often implemented as tiny 1x1pixel images on web pages or emails. When the page loads, the image is downloaded from the server, creating a record on the server logs. Page tags are often used in email campaigns such as newsletters.

Issues: Emails opened as plain text (or text-based browsers) do not download this tracked image and thus may impair reporting.

User Agent String

The user agent string is information which that can be requested by a script or server, that reveals the browser and operating system details of the user accessing that page.

Issues: The user agent string can be easily faked. Firefox even has an add-on to switch your default user agent on the fly.

Explanation of the user agent string from whatsmyuseragent.com

Collecting Accurate Data

As there are a number of issues with each type of data collection method above, a more accurate way of recording interaction with your site would be to incorporate a number of different collection methods in one go.

Google Analytics uses a combination of factors like these in order to more accurately reflect a website’s usage.

For example: Capturing a visitor’s IP address will tell us the information provided by their router. If we also have the user agent string from their browser, this tells us their browser type and the computer’s operating system. We now have a much better idea as to whether or not we are getting multiple visits from one person or a series of different users that are all on the same network.