Learning how our web site works is an important task, which is often measured incorrectly. We have to look at not only the data, but we also need to know what questions to ask when looking at the data to put it into the correct perspective.
Without understanding how the data works, we can run into all types of problems by drawing the wrong conclusions.
In this section, I assume that we will be looking at data from some source such as Google Analytics and we are wanting people to purchase something from us. Not all web sites are built with purchasing in mind, some want to allow answers to be found, collect user registrations, etc. We will group all of these into a generic “end goal” that you have predetermined. It is important to know what your web site is supposed to do, so you know this end-goal. Is it selling products, providing information, getting someone to arrange an appointment with you, or letting people find answers on their own so they don’t have to call you?
Another key thing to note, is that your metrics information is going to have flaws. There are a wide variety of reasons for this, but you must know and be willing to accept this. We must be willing to assume a certain level reporting errors in our data. While I cannot go into all of the reasons why there might be errors, I will try to point out some obvious reasons why there might be errors as we go along the process.
The obvious measurement
There are a couple of “obvious” measurements. Everyone looks at these because they are often predominately displayed in your analytics tool. They would be:
- Visits,
- Page Views, and
- Unique Visitors.
A Visit is one of the two most commonly looked at pieces of information which people look at. It is also one of the most useless pieces of information when it is looked at by itself. A visit is when a user goes to your site. Their entire experience there, be viewing only one page, or 100 pages, is summed up as a single visit. It generally assumes that this is an active process, and the user does not break for more than a few minutes between any two pages. If the user leaves the site, and returns the next day, or even an hour later, this is a second visit.
Why it is important to know: Without visitors, your web site sits idle and unproductive. By measuring visitors you can start to track other, more important, pieces of information. namely how many people are completing their “final goal”. By using the number of visits your web site has, vs. the number of final goals completed, you can start to determine what your conversion rate is.
Why it can be misleading: There are a couple of things that can get in the way of you understanding your visits metric. First, not every visitor is tracked. This is usually for technical reasons beyond your control. For example, if your software relies on use of a JavaScript, your visitor might have JavaScript disabled. If you have to send a cookie to a third party, these can be blocked by your web browser, of if it looks at log files from your web server, your pages might be cached and accessed by a Internet Service Provider. Each of these, and other reasons, are beyond your control, and can effect the quality of your numbers. You have to be willing to live with these errors, and work to establish the best information you can even with these problems, knowing that you don’t have 100% accurate data to work with.
Of course if there is a metric which someone loves more than Visitors, it has to be Page Views. This is approximately, see “Why it can be misleading” for visitor counts, how many web pages have been viewed by users. People love this number because it is often “big”. However, this can be the most misleading metric of all time. Why; because Page Views mean nothing by themselves. Page views, unless you are using an advertising based model, mean and show nothing. However, at the same time, Page Views are extremely important. For without them, you cannot move someone to the end goal. But you must remember that they are not the end goal.
Why it can be misleading: Number one, psychology. We love to see this number go up, and it can cause us to look solely at this number, without seeing how the end goal is affected. We cannot be distracted by the sirens call, and so the first problem with this metric is we cannot focus on it. Second, this number is often artificially low. Depending upon how your metric tool works, page reloads/revisits may not be calculated as they could be cached either by your end user’s computer and/or their web host to reduce the amount of bandwidth they use.
A Unique Visitor is someone who has gone to your web site. If they visit your site five times throughout the reporting period (day, week, month, etc), they should only be counted once.
Why it is important to know: By understanding how many unique people have visited our web site, we can start to determine if we are providing a site which users want/need to return to. Many web sites are built with this purpose. As people return to our site, we develop a sense of authority with this person, which can allow them to return, and/or encourage them to complete the end-goal.
Why it can be misleading: This number might be inaccurate however because we must track a person to a computer. So if I were to visit a site twice while at work, once on a mobile phone while commuting home, and twice later that evening, it would show as three unique visitors because I used three devices.
It can also be wrong because of technical reasons. For example: the cookie, which was used to track the user, was deleted between visits. Now the device, which was used to visit the site, appears to not have visited before, and we show a second unique user when in fact only one exist.
Once we realize that our data will never be complete, that is OK. We can move forward a lot easier knowing we are 70-90% accurate, and just have to make decisions with that in mind.
In our next article we will look at how many pages your visitors see, as well as how long they are on your site.
About Walter Wimberly