Advanced Statistics

 Objective - Problematics
The Internet is anonymous by nature. It is not possible to know the names of the visitors to your site without explicitly requesting this information (by restricting component access to certain users, for example).

The visit meter, counting only the visits that come in through the home page, is a simplistic and obsolete tool when compared to the advanced statistics application. Indeed, with the visit meter it is impossible to know the dates of these visits, or the number and nature of the viewed pages.

Let's first clarify the definition of a 'visit'. By nature, Internet browsing does not limit a visitor's possibilities. The visitor can leave your site at any time and come back if desired. When a visitor comes to your site, our servers open a work session with them. They send their browser a piece of information that it stores as long as the user doesn't close their browser and for at least fifteen minutes after their last interaction with your site.

To put it more clearly, this means that if the visitor comes to your site, clicks on a link directing them to another site, and then comes back to your site (within fifteen minutes), then we consider that they are still in the same visit.

This also means that if a user begins to fill their shopping cart in your site and they get a phone call, the user has fifteen minutes to resume their work before our servers close their work session.

This fifteen minute time period is arbitrary and is based on the experience of our technicians.

On the Internet, nothing obliges a user to indicate that they have left your site or allows you to know if they just left your site to go and visit another or if something in everyday life (a telephone call, a conversation, etc.) has interrupted their visit.

During a work session (what we call a 'visit' in the rest of our text), the visitor is going to view different pages of your site.

Contrary to the visit meter that is going to give you a count at a given moment, the advanced statistics module is going to allow you to understand what your users are doing and possibly use this information to modify your site in order to improve the productivity of these visits.

Let's examine the different available information.
 Table of viewed pages for a given period
To obtain this data, select the "Display the details of the viewed pages for a given period" mode and choose an analysis period, for example "Last 7 days".

Click on the button "Apply Filter".

By default, the displayed table will be sorted by the number of viewed pages. You can however change this order to obtain a classification by component or by the date of the last visit.

The tables are displayed with the component name, the date of the last visit, and the number of viewed pages.

In the case where the analysis period is spread out over several days, a small "+" icon will be displayed to the right of the number of viewed pages for each component and will allow you to consult the details of the viewed pages for this component and for each day in the period.
 Table of viewed pages for a given period (detailed mode)
Now select the "Display the details of the viewed pages for a given period (detailed mode)" mode.

The table is no longer detailed solely by component, but is automatically detailed by component and by view.

Each component can have one or several views. Each view represents a type of displayed page.

As the name indicates, the "default" view represents the main default view (such as the display page for the products in the catalog).

Let's take the shopping cart component for example. This component is organized around 8 different views:

View Signification
Default Product display pages.
Search Search pages.
Shopping Cart Shopping cart page where the buyer indicates quantity.
Order file Page where the buyer enters their information (address, phone number, etc.).
Shipping Options Page containing the shipping options.
Payment Methods Page containing the payment method choices.
Confirmation Order confirmation page.
Thank You Thank you page at the end of the order.

This 'view' concept is used in other components, such as the FAQ, where, in addition to the default view which displays the questions and answers, there is a redaction form that corresponds to the question redaction page.

Let's take a case in point:

View Signification Viewed Pages
Default Product display pages. 24 826
Search Search pages. 335
Shopping Cart Shopping cart page where the buyer indicates quantity. 1 827
Order File Page where the buyer enters their information. 253
Shipping Options Page containing the shipping option. 225
Payment Methods Page containing the payment method choices. 182
Confirmation Order confirmation page. 189
Thank You Thank you page at the end of the order. 120

This signifies that your visitors have viewed 24,826 pages of your catalog, that they have used the search function and viewed 335 search result pages, that they have added to (or modified) the shopping cart 1827 times, and that they have come to the information entry form 253 times.

Basically, this signifies that 253 people had the intention to buy (knowing that if someone does not correctly enter their data and that the system displays the entry page for them again, this counts as another display. Therefore, this figure (253) should not be seen as an exact reference but as a very good approximation.

Then, 225 people came to the shipping options page. This means that certain people abandoned their purchase and that others may not be concerned by shipping (e-product sales, free shipping, etc.).

Finally, there were 182 displays of the payment page. Therefore, about 182 people decided to continue.

However, the system indicates that the order confirmation page was displayed 189 times. Don't take this to be an error because only 182 people passed through the previous page. These are not people, but page displays! This simply means that a small proportion of users refreshed this page, returned to the shopping cart, etc.

In the end, 120 people saw the thank you message. This clearly means that certain customers did not follow through with payment, that certain others paid but did not return to your shop at the end of payment to see your thank you message, etc.

The analysis of this case allows us to observe the purely statistical character of this tool.

For example, analyzing the evolution of the ratio 1827/24826 after a period of time will allow you to acquire an understanding of how attractive your products and your prices are. The higher the rate is, the more this signifies that people have the tendency to want to add the products to the cart.

The ratio Shopping Cart View/Order File View doesn't make any sense because you may have orders with a high volume of products and others that contain very few.

The ratio Order File/Shipping Options is meaningful because it allows you to observe if your shipping fees are possibly too high.

The other ratios can also provide you with an abundant amount of information.
 Line graphs of visits and viewed pages
Now select "Display a line graph of the visits and viewed pages".

Note: A line will not be displayed unless at least two values are available. Therefore, there won't be any results for analysis periods of one day (yesterday, day before yesterday, etc.).

Two lines will then be displayed, one reflecting the visits and the other reflecting the viewed pages.

By hovering over the graph legend with your mouse, the value details will be displayed.
 Table of the origin details
Finally, select "Display the origin details".

Like a large proportion of the Internet, the advanced statistics are declarative and are, therefore, based on the "declaration of the sending server".


You possess the site and you create a link from your site to the site A visitor surfs on the site and clicks on the link to visit your site At this precise moment, our servers are responsible for knowing where this new visitor, coming to your site, is coming from. Consequently, our servers ask the visitor's browser "Where were you before?". If the answer is, then our servers can record this intelligible answer for the statistics.

However, if the server of the site decides to answer "" (an IP address), which is their absolute right, then our servers are going to conclude "origin unknown, answer unintelligible".

Certain engines use this method to avoid revealing their true impact on the traffic of your site.

In the same way, if you open a browser and you directly type in the address bar, the sender source is the IP address that you are using to surf the net and that also enters into the "unknown and unintelligible" category.

This has prompted us to introduce another Actinic service: ad campaign tracking. As the example above has demonstrated, it is difficult to trust the sender to determine the exact origin of the visit, particularly when you pay to obtain this visit (traffic generation, sponsored key words, price engines, etc.).

In such a case, the origin statistics are not the best tool. The campaign tracking system gets around this problem by creating "fake pages" in your site that you ask your traffic suppliers to "call". By doing this instead of asking yourself where the customer comes from to know where they came from [sic], knowing that this information may be concealed, the system deduces the origin in relation to where the customer arrives (because you will give a different "false page" to each supplier).

Click on the icon for more details about the campaign tracking system:

 You want to test if its works?
No problem! It's natural to want to understand how it all works, but be careful! Here are several traps to avoid!

First of all, understand that when you preview your site, your visits are not included in your statistics. If this were the case, each time you wanted to test your site, it would be counted as a visit!

Therefore, if the name of your site is 'YourSite', open a new browser and go directly to in the same way a visitor would! Don't use the preview buttons that are available in your site's administration space.

Visit your site. Browse through your pages, add elements to your shopping cart, ask a question in the FAQ component, and sign the guest book! In short, behave in the same way as an average visitor would. Then close your browser window.

Now wait ... until the following day! Indeed, even though the statistics are recorded at the time of viewing, the colossal amount of data produced does not allow for a direct analysis. Every night our servers compile this information to make it easily accessible and to allow you to perform a fine-tuned analysis while, at the same time, keeping costs down.

 Deleted pages:
It is possible that a particular page that you have deleted from your site (WebBlock, HTML, File) is still being indexed by a search engine or saved in the favorites of one of your visitors.

It is, therefore, possible that someone may attempt to visit this page. Naturally, it is impossible to display a deleted page or to even find its name after it has been deleted. Consequently, the statistics system will display a red line, such as "[12345]? (WebBlock)" to inform you that an attempt has been made to visit a nonexistent page. The number 12345 used here is just an example. You will have different numbers that correspond to unfound WebBlock, HTML page and file identifiers. Apart from informing you of the visit attempt, we cannot find the elements that you have deleted and that are still referenced on the Internet. Don't pay too much attention to these red lines unless a large number of viewed pages are attached to these errors. In this case, browse through your site and check the different links in the WebBlocks, HTML pages and uploaded files to try and locate the cause of the error, which may be coming from a link to a deleted page or WebBlock that still exists in your site.


Add to favorites:
© 2001-2018 Oxatis. All rights reserved.