Web Analysis is a set of techniques related to the analysis of data related to traffic in a website with the aim of understanding its traffic as a starting point to optimize various aspects of it.
There are two categories of web analysis: off-site and on-site.
Off-site web analysis refers to web measurements and independent analyzes of whether the website owner is or is being maintained. This includes measuring potential audience (opportunity), voice participation (visibility) and buzzing (comments) about what is happening on the Internet as a whole.
On-site web analysis measure a visitor's path once they enter a web site owned by the person performing the analysis. This includes conversions; for example what pages of arrival encourage people to make a purchase. On-site web analytics measurements compare key performance indicators and use it to improve a website or audience response to a marketing campaign.
Historically, web analysis has made reference to the measurement of on-site visitors. However, in recent years there has been a convergence of both aspects, mainly because vendors are producing tools that cover both categories.
The sources of information in Web Analysis are:
Web Analysis tools.
Online Advertising Servers.
Email marketing tools (mass mailing).
Corporate databases, customers, suppliers, etc.
Web analysis tools are responsible for capturing and processing website information to provide information about users' behavior on the site: the site from which they come, what they do on the site, what pages they browse, for how long time, how often they revisit the site, what country they are, what type of internet connection they have, where they leave the site, what step of a process they discontinue, etc.
Web Analysis tools can be based on different technology platforms:
Log file analyzers
They are programs that analyze the logs of the servers providing information about "who", "when" and "how" they visit them.
Its main advantages are:
- Servers always produce log files, so information is always available.
- The servers capture all access to the site.
- The information normally resides on the servers themselves and has a standardized format. This facilitates the migration of some tools to others.
- The log files store information about the failed requests, while with other techniques, it is lost.
This method, more recent than the previous one, is based on the incorporation of a script to each of the pages of a site. Each time a page is visited, this script communicates with a database to which it communicates the printing of the page together with, potentially, additional data coming from the cookies.
This technique has the following advantages:
Potentially, you can capture information not available in the log files and even modify it without changing the scripts.
The labeling of pages can be done even in cases where the owners of the site do not have access (for being hosted on other servers, for example) to the server logs.
Hybrid systems Some companies have developed solutions that combine both solutions adding the individual advantages of the same ones.
Packet Sniffing The packet sniffer is added between the user's computer and the site server, so it has an optimal information capturing capability, capturing all the information that is generated, being a Page View or not.
Advantages The main advantage of packet sniffing in terms of data collection is the fact that all the information is captured, whether or not a Pageview has been generated, whether the contents are downloaded or not.
- High performance.
- Easy implementation.
- No need to set anything to analyze the clickstream.
Economic factors The analysis of the log files is usually done internally. The acquisition of the necessary software requires an initial single outlay, although there are also excellent free analyzers.
However, the analysis of data from the labeling of pages is usually (subcontracted) to companies that may require periodic payments depending on the level of service.
The most appropriate alternative depends on the internal technical knowledge of the site owner, the depth of the analysis required, the provider, the volume of traffic, etc.
There are no globally approved definitions of web analytics even though industry agencies have been trying to agree on definitions that are useful and definitive for some time. The main organizations that have contributed in this area have been Jicwebs (Industry Committee for Web Standards) / ABCe (Auditing Bureau of Circulations electronic, UK and Europe), WAA (Web Analytics Association, US) and in a smaller proportion IAB Advertising Bureau)
Most common definitions presented by WAA and ABCe:
Hit: A request for a file to the web server. It is available only in analysis logs. The number of hits received by a website is frequently quoted to deduce its popularity, but this number is extremely misleading and dramatically overestimates popularity. A single web page usually consists of multiple (sometimes dozens) files, each of which is counted as one hit even though a single page is downloaded, so the number of hits is actually an arbitrary number rather than the number Complexity of individual pages on the website make the actual popularity of the website. The total number of visitors or page views provides a more realistic and accurate appreciation of popularity.
Page view: A request for a file whose type is defined as a page in the analysis log. In that log, the single page view can generate multiple hits as well as all the resources required to view the page (images, .js and .css files) are also requested by the web server.
Visit / Session: A series of requests from the same client uniquely identified in a range of time, sometimes 30 minutes. A visitor contains one or more page views.
First visit / First session: A visit from a user who has not previously entered the site.
Visitor / Unique Visitor / Unique User: The client uniquely identified by generating requests to the website (log analysis) or viewing pages (page tagging) in a given period (eg day, week or month). A unique visitor counts only once on the time scale. A visitor can make multiple visits. The identification is made with respect to the visitor's computer, not the person, usually via cookies and / or IP + User Agent. That is why the same person visiting from two different computers will count as two unique visitors.
Repeated visitor: A visitor who has arrived at the site at a previous opportunity. The period between your last visit and the current visit is measured in days.
New visitor: A visitor who has not made any previous visits. This definition creates a certain level of confusion, and is sometimes replaced with analysis of first visits.
Print: It's every time an ad loads on a user's screen. Every time you see a notice, that's an impression.
Singleton: The number of visits where only one page is viewed. While this is not a useful metric by itself, the number of singletons is indicative of various forms of click fraud as well as being used to calculate the dropout rate and in some cases to identify robot robots.
Bounce rate: The percentage of visits where the visitor enters and leaves the same page without visiting other pages on the site.
Output percentage: The percentage of users leaving a page.
Visibility time: The time that on a single page (or a blog, or advertisement) is viewed.
Session duration: The average amount of time that visitors spend on the site each time they visit. This metric can be complicated by the fact that analysis programs can not measure the duration of the last page view.
Duration of page view / Time on page: Amount of average time that visitors spend on each page of the site. As with the duration of the session, this metric is complicated by the fact that analysis programs can not measure the duration of the last page view.
Page Depth / Pageviews Per Session: The page depth is the average number of pageviews a visitor consumes before finishing their session. It is calculated by dividing the total number of page views between the total number of sessions, and is also called pages viewed per session.
Frequency / Single Session - Frequency measures how often visitors arrive at a site. It is calculated by dividing the total number of sessions (or visits) between the number
You May Also Like