Web Analytics is the measurement, collection, analysis and reporting of web data for the purpose of understanding and optimizing web usage. However, Web analytics is not just a process for measuring web traffic but can be used as a tool for business and market research, and for assessing and improving website effectiveness. Web analytics applications can also help companies measure the results of traditional print or broadcast ad campaigns. This helps one to estimate how traffic to the website changes after the launch of a new advertising campaign. Web analytics provides information about the number of visitors to the website and the number of page views. It helps measure traffic trends and popularity that are useful for market research.
Video Web analytics
Basic steps of the web analytics process
Much of the web analytics process goes down to four important stages or steps:
- Data collection: This stage is the basic, basic data set. Typically, this data is the number of things. The purpose of this stage is to collect data.
- Process data into information: This stage usually takes the number and makes them the ratio, although there may still be some counts. The purpose of this stage is to capture the data and customize it into information, especially metrics.
- Developing KPIs: This stage focuses on the use of ratios (and numbers) and instilling them with business strategies, called the Main Performance Indicators (KPI). Many times, KPI deals with conversion aspects, but not always. It depends on the organization.
- Formulate an online strategy: This stage deals with online goals, goals, and standards for organizations or businesses. This strategy is usually associated with making money, saving, or increasing market share.
Another important function developed by analysts for website optimization is experimentation
- Experiments and testing: A/B testing is a controlled experiment with two variants, in an online setting, such as web development.
The purpose of A/B testing is to identify changes to web pages that improve or maximize statistically tested results.
Each stage impacts or may affect (ie, encourage) the previous stages or follow it. So sometimes the data available for the collection impacts on the online strategy. At other times, the online strategy affects the data collected.
Web analytics technology
There are at least two categories of web analytics; off-site and on-site web analytics.
Historically, web analytics has been used to refer to visitor measurements in place. However, this meaning becomes blurred, especially since vendors produce tools that span these two categories. Many different vendors provide web analytics software and services on the spot. There are two main technical ways to collect data. The first and traditional method, server log file analysis , reads the files in which the web server records the file requests by the browser. The second method, page tagging , uses JavaScript embedded in web pages to create image requests to third-party custom analytics servers, each time a web page is rendered by a web browser or, if desired, when a mouse click occurs. Both collect data that can be processed to generate web traffic reports.
Web data analysis source
The fundamental purpose of web analytics is to collect and analyze data related to web traffic and usage patterns. The main data comes from four sources:
- HTTP direct request data: directly from an HTTP request message (HTTP request header).
- The server-generated network and data levels associated with the HTTP request: are not part of the HTTP request, but are required for successful request transmission. For example, the applicant's IP address.
- Application-level data sent with HTTP requests: created and processed by application-level programs (such as JavaScript, PHP, and ASP.Net), including sessions and referrals. This is usually captured by internal logs rather than public web analytics services.
- External data: can be combined with on-site data to help add the website behavior data described above and interpret web usage. For example, IP addresses are typically associated with geographic areas and internet service providers, open email and clickthrough rates, direct mail campaign data, sales history and leads, or other data types as needed.
Analysis of web server log files
The web server records some of their transactions in the log file. Immediately realized that these log files can be read by a program to provide data about the popularity of the website. So comes web log analysis software.
In the early 1990s, website statistics consisted mainly of counting the number of client requests (or hits ) made to the web server. This is a reasonable method at first, because every website often consists of one HTML file. However, with the introduction of images in HTML, and websites that span multiple HTML files, this count becomes less useful. The first true commercial Analyzer Log released by IPRO in 1994.
Two units of measure were introduced in the mid-1990s to measure more accurately the amount of human activity on a web server. These are page views and visits (or sessions ). The page view is defined as a request made to the web server for the page, as opposed to the graph, while the visit is defined as the query sequence of a recognized clientele that expires after inactivity in the amount certain, usually 30 minutes. Page views and visits are still displayed metrics frequently, but are now considered somewhat incomplete.
The emergence of search engine spiders and robots in the late 1990s, along with web proxies and dynamically assigned IP addresses for large corporations and ISPs, made it more difficult to identify unique human visitors to websites. The log analyzer responds by tracking visits by cookies, and by ignoring requests from known spiders.
The widespread use of web cache also presents a problem for log file analysis. If a person revisits a page, a second request will often be cached from the browser cache, so no request will be accepted by the web server. This means that the person's path through the site is lost. Caching can be defeated by configuring a web server, but this can result in decreased performance for visitors and greater load on the server.
Page tagging
Concerns about the accuracy of log file analysis in the presence of caching, and the desire to be able to perform web analytics as an outsourced service, lead to second data collection methods, page tagging or 'Web bugs'.
In the mid-1990s, Web counters were often seen - this is a picture entered in a web page that shows how many times it was requested, which is the approximate number of visits to the page. In the late 1990s this concept evolved to include a small image that was invisible rather than visible, and, using JavaScript, to forward with image requests for certain information about pages and visitors. This information can then be processed remotely by the web analytics company, and the extensive statistics generated.
The web analytics service also manages the process of providing cookies to users, who can uniquely identify them during their visits and in subsequent visits. The acceptance level of cookies varies significantly between websites and may affect the quality of data collected and reported.
Collecting website data using a third-party data collection server (or even an in-house data collection server) requires additional DNS search by the user's computer to determine the IP address of the collection server. Sometimes, delays in completing a successful or failed DNS search may cause data not to be collected.
With the increasing popularity of Ajax-based solutions, an alternative to the use of unseen images is to apply the call back to the server from the given page. In this case, when a page is displayed in a web browser, a collection of Ajax code will call back to the server and provide information about the client which can then be aggregated by the web analytics company. This is in some way disabled by browser restrictions on the server that can be reached with the XmlHttpRequest object. Also, this method can cause the reported traffic levels to be slightly lower, as visitors can stop the page from loading in mid-response before an Ajax call is made.
Logfile analysis vs. page tagging
Both logfile analysis programs and page tagging solutions are available for companies looking to do web analytics. In some cases, the same web analytics company will offer both approaches. The question then arises from which method should the company choose. There are advantages and disadvantages of each approach.
Advantages of logfile analysis
The main advantages of log file analysis over page tagging are as follows:
- The web server usually generates log files, so raw data is readily available. No changes to website required.
- The data resides on a company's own server, and in a standard format, not proprietary. This makes it easier for companies to switch programs later, use several different programs, and analyze historical data with new programs.
- The log file contains information about visits from search engine spiders, which generally do not execute JavaScript on the page and therefore are not logged by page tagging. While this should not be reported as part of human activity, it is useful information for search engine optimization.
- Logfile does not require additional DNS lookup or slow startup of TCP. So no external server calls can slow down the page loading speed, or generate innumerable page views.
- The web server can reliably record every transaction made, e.g. serving PDF documents and script-generated content, and not relying on a visitor's browser that works together.
The benefits of tagging a page
The main advantages of tagging the top pages of log file analysis are as follows:
- The calculation is enabled by opening the page (remembering that the web client runs the tag script), does not request it from the server. If the page is cached, it will not be counted by server-based log analysis. Cached pages can reach up to one third of all page views. Not counting the cached pages seriously alters many of the site's metrics. It is for this reason that server-based analysis logs are not considered suitable for the analysis of human activity on the website.
- Data is collected through a component ("tag") on a page, usually written in JavaScript, although Java can be used, and more Flash is used. Ajax can also be used in conjunction with a server-side scripting language (such as PHP) to manipulate and (typically) store it in a database, essentially allowing full control over how data is represented.
- Scripts may have access to additional information on web clients or users, not sent in queries, such as visitor screen sizes and prices of items they buy.
- Pages can report events that do not involve requests to web servers, such as interactions in Flash movies, partial completion of forms, mouse events like onClick, onMouseOver, onFocus, onBlur etc.
- The page tagging service governs the cookie-giving process to visitors; with log file analysis, the server must be configured to do this.
- Paging is available for companies that do not have access to their own web servers.
- Lately tagging pages has become the standard in web analytics.
Economic factors
Logfile analysis is almost always done at home. Page tagging can be done in-house, but more often provided as a third-party service. The economic difference between the two models can also be a consideration for companies that decide to buy.
- Logfile analysis typically involves purchasing a one-time software; however, some vendors introduce maximum annual page views with additional costs for processing additional information. In addition to commercial offerings, some open source logfile analysis tools are available for free.
- For Logfile analysis you have to store and archive your own data, which often grows very quickly quickly. Although the cost of hardware to do this is very minimal, the additional cost to the IT department can be overwhelming.
- For Logfile analysis, you need to maintain the software, including updates and security patches.
- The complex page marking vendors charge a monthly fee based on volume that is the number of page views per month collected.
Which cheaper solution to apply depends on the amount of technical expertise within the company, the vendor selected, the number of activity seen on the website, the depth and type of information sought, and the number of different websites that require statistics.
Regardless of the vendor solution or data collection method used, the cost of analysis and interpretation of web visitors should also be included. That is, the cost of converting raw data into actionable information. This can come from the use of a third-party consultant, hiring an experienced web analyst, or training from the appropriate home-based person. A cost-benefit analysis can then be done. For example, what revenue or cost savings can be gained by analyzing web visitor data?
Hybrid method
Some companies produce solutions that collect data through both log files and page tagging and can analyze both types. By using hybrid methods, they aim to produce more accurate statistics than any other method. The initial hybrid solution was produced in 1998 by Rufus Evison.
Geolocation of visitors
With IP geolocation, it is possible to track the visitor's location. By using an IP or API geolocation database, visitors may be located geolocation to city, region or country level.
IP Intelligence, or Internet Protocol Intelligence (IP), is a technology that maps the Internet and categorizes IP addresses based on parameters such as geographic location (country, region, state, city and zip code), connection type, Internet Service Provider (ISP) proxies, and more. The first generation of IP Intelligence is referred to as geotargeting or geolocation technology. This information is used by businesses for in-app online audience segmentation such as online advertising, behavioral targeting, content localization (or website localization), digital rights management, personalization, online fraud detection, localized searches, improved analysis, global traffic management, and content distribution.
Analytics click
Click analysis is a special type of web analytics that pays special attention to clicks.
Generally, click analytics focuses on analytics in place. The website editor uses click analytics to determine the performance of the site in particular, with regard to where the site's users clicked.
Also, click analytics can occur in real-time or "unreal" depending on the type of information sought. Typically, the front page editor on high traffic media news sites will want to monitor their pages in real-time, to optimize the content. Editor, designer, or other type of stakeholder can analyze clicks over a broader timeframe to help them assess the writer's performance, design elements, or advertising etc.
Data on clicks can be collected in at least two ways. Ideally, a click is "logged in" when it happens, and this method requires some function that retrieves relevant information when the event occurs. Alternatively, one may institute the assumption that a page view is the result of a click, and therefore a simulated click log that leads to the page view.
Customer lifecycle analysis
The customer life cycle analysis is a visitor-centered approach to measuring that is under the marketing lifecycle umbrella. Pageviews, clicks and other events (such as API calls, access to third-party services, etc.) are all related to individual visitors instead of being stored as separate data points. The customer lifecycle analysis seeks to connect all data points into marketing funnels that can offer insight into visitor behavior and website optimization.
Other methods
Other data collection methods are sometimes used. Sniff packets collect data by sniffing network traffic passing between the web server and the outside world. The sniffing package does not involve changes to web pages or web servers. Integrating web analytics into the web server software itself is also possible. Both of these methods claim to provide better real-time data than other methods.
Maps Web analytics
Web analytics in place - definition
Off-site web analysis is based on open data analysis, social media exploration, sound distribution of web properties. Usually used to understand how to market your site by identifying keywords that are tagged to your site, whether from social media or from other websites.
By using HTTP Referer, the owner of the web page will be able to track which one is the referring site that helps bring traffic to their own site.
Common sources of confusion in web analytics
Hotel problem
The problem of hotels is generally the first problem faced by web analytics users. The problem is that unique visitors for each day of the month do not add up to the same number of unique visitors for that month. This appears to be an inexperienced user a problem in whatever analytics software they use. Actually it is a simple property of the metric definition.
How to describe the situation is to imagine a hotel. The hotel has two rooms (Room A and Room B).
As pointed out by the table, the hotel has two unique users every day for three days. The sum of the total with respect to the day is six.
During that period each room has two unique users. The sum of the total with respect to the rooms is four.
Actually only three visitors have been in the hotel during this period. The problem is that someone staying in a room for two nights will count twice if you count them once every day, but only count once if you look at the total for that period. Any software for web analytics will sum this up correctly for the selected time period, leading to problems when users try to compare totals.
Web analytics method
Historically, the page tagging analysis analytics vendor has used third-party cookies sent from the vendor's domain, not the domain of the browsed website. Third-party cookies can handle visitors who cross multiple unrelated domains within the company's site, since cookies are always handled by the vendor's servers.
However, third-party cookies in principle enable individual user tracking across different company sites, enabling analytics vendors to structure user activity on sites where they provide personal information with their activity on other sites that he thinks are anonymous. Although web analytics companies refuse to do this, other companies such as companies that provide banner ads have done so. Privacy concerns about cookies have caused a visible minority of users to block or delete third-party cookies. In 2005, some reports indicate that about 28% of internet users block third-party cookies and 22% delete them at least once a month. Most page-tagging solutions vendors have now been moved to provide at least the option of using first-party cookies (cookies provided from client subdomains).
Another problem is the deletion of cookies. When web analytics relies on cookies to identify unique visitors, the statistics depend on a fixed cookie to hold unique visitor IDs. When a user deletes a cookie, they usually delete the first and third party cookies. If this is done between interactions with the site, the user will appear as a first-time visitor at the next interaction point. Without a fixed and unique visitor identity, conversions, click-flow analysis, and other metrics that depend on unique visitor activity over time, can not be accurate.
Cookies are used because IP addresses are not always unique to users and can be shared by large groups or proxies. In some cases, IP addresses are merged with user agents to more accurately identify visitors if cookies are not available. However, this only resolves part of the problem because often the user behind the proxy server has the same user agent. Other methods of identifying users are technically technically challenging and will limit the audience that can be tracked or will be considered suspicious. Cookies are the choice chosen because they reach the lowest common denominator without using technology that is considered spyware.
Metering method
It may be good to realize that third party information collection is subject to the network and security restrictions being applied. Countries, Service Providers, and Private Networks may prevent your site traffic data from going to third parties. All of the methods described above (and some other methods not mentioned here, like sampling) have major problems that are vulnerable to manipulation (both inflation and deflation). This means this method is not appropriate and unsafe (in a reasonable security model). This issue has been discussed in a number of papers, but to-date solutions suggested in this paper remain theoretical, perhaps due to a lack of interest from the engineering community, or because of the financial benefits the current situation provides to the owners of large websites. For more details, read the above mentioned papers.
See also
References
Bibliography
- Clifton, Brian (2010) Advanced Web Metrics with Google Analytics, second edition, Sybex (Paperback.)
- Kaushik, Avinash (2009) Web Analytics 2.0 - Online Responsibility Art and Customer Care Concern. Sybex, Wiley.
- Mortensen, Dennis R. (2009) Yahoo! Web Analytics. Sybex.
- Farris, P., Bendle, N.T., Pfeifer, P.E. Reibstein, D.J. (2009) Main Marketing Metrics 50 Metrics Every Manager should know, Prentice Hall, London.
- Plaza, B (2009) Monitor the effectiveness of web traffic sources with Google Analytics: Experiments with time series. Aslib Proceedings , 61 (5): 474-482.
- Arikan, Akin (2008) Multichannel Marketing. Metrics and Methods for Active and Offline Success. Sybex.
- Tullis, Tom & amp; Albert, Bill (2008) Measuring User Experience. Collecting, Analyzing, and Presenting Metrics Usability. Morgan Kaufmann, Elsevier, Burlington MA.
- Kaushik, Avinash (2007), Web Analysis: One Hour a Day, Sybex, Wiley.
- Bradley N (2007) Marketing Research. Tools and Techniques. Oxford University Press, Oxford.
- Burby, Jason, and Atchison, Shane (2007) Actionable Web Analysis: Using Data to Make Smart Business Decisions.
- Davis, J. (2006) "Marketing Metrics: How to create a truly Accountable Marketing plan 'John Wiley & Sons (Asia).
- Peterson Eric T (2005) Measurement of Website Hacks. Ebook O'Reilly.
- Peterson Eric T (2004) Web Analytics Demystified: A Marketer's Guide to Understanding How Your Website Affects Your Business. Celilo Group Media
- Lenskold, J. (2003) 'Marketing ROI: how to Plan, Measure and Optimize Profit strategies' London: McGraw Hill Contemporary
- Sterne, J. (2002) Web Metrics, Proven Methods for Measuring Website Success, London: John Wiley & amp; Boys.
- Srinivasan, J. (2001) E commerce Metrics, Models and Examples, London: Prentice Hall.
- Zheng, J. G. and Peltsverger, S. (2015) Overview of Web Analytics, In the book: Encyclopedia of Science and Information Technology, Third Edition, Publisher: IGI Global, Editor: Mehdi Khosrow-Pour
Source of the article : Wikipedia
0 Comments