How Accurate is your Web Analytics Data
- Monday, April 20, 2009
To be able to judge your website performance, your web analytics should be accurate. You will be able to achieve this if common errors in collecting the data can be avoided. You need to understand your website visitors behavior along with their entry and exit points.
In today’s world Web Analytics are the tools and methods used by organizations to track the number of people visiting their website and later analyzing the success of their online strategies. Web analytics data can never be 100% accurate. Finding the errors is difficult and the levels of details are critical.
Collecting Visitors Data
The two common techniques to collect the website visitors data are the Log Files data and the Page Tags (scripts). Log Files data comes from your web server and independent of the visitors browser. It is also known as “server-side data collection", it captures all the requests made by the website visitors to your web server including the visitors IP address, date/time, web pages, images and PDFs.
The best part about log files are they can track mobile visitors also as most of the mobile phones yet do not support Java Script or cookies. Log files are able to track all visitors who browse using their mobile phones, which a Page Tags can’t do. Log files can also track all the hits to your website PDF’s, DOC’s, EXE’s, ZIP’s & XLS’s files which a Page Tag can’t do as you cannot tag these files.
But the visits by Robots/Spiders/Crawlers can also multiply your data figures, as search engines use them to fetch and index your website. These affect web analytics because log file will show all data for robots activity on your website even though they are not real visitors. In Log Files, one IP address counts as one visitor. A log file tracks visitor sessions by accumulating all hits from the same IP address and web browser per visitor.
It is a problem when ISPs (internet service providers) assign different IP addresses in whole session, in that case the same visitors will be counted multiple times. As a result, the visitors’ numbers are over-counted and become very high. This can be overcome by the proper use of cookies. Page Tags collect website visitors data via visitors web browsers, this information is normally captured by Java Script codes placed on each and every page of the website. It is also known as “client-side data collection", and this is the most commonly used technique by Web Masters and Vendors.
It is always recommended that the Page Tags should be placed at the bottom of the page just above the </body> tag of the HTML web page. When the web page has loaded completely, this becomes a valid hit/count. The potential problem here is that the repeated visitors who are very familiar with your website navigation may navigate quickly by clicking on to another page before the Page Tag has loaded to collect data. Page Tags are very helpful and can track each and every move of the website visitor, but it can also have setup errors as it relies on the web masters/developers to add those tags on each and every page. Problem rises if some pages get missed and those missing tags mean loss of data for those page views.
Java Script Page Tags work well provided Java Script is enabled on the visitors web browsers. Also any error in other Java Script code on the same web pages can immediately stop the browser scripting engine at that point, therefore a Page Tag placed below it will not execute. Corporate and personal firewalls that can prevent Page Tags from sending the data to the collection server. Firewalls can also be set up to reject or delete cookies automatically, which can significantly affect on the visitors data. Imagine a person accessing your website from work, home and Internet cafes, this will result in three different cookies.
All web analytics solutions will count each of these anonymous user sessions as unique visitors. On the other hand people share their computers all the time, particularly with their families & friends and as a result cookies are shared too. Summary Web Analytics is not 100% accurate and the numbers of inaccuracies are likely to happen. This is a universal truth. So be comfortable with your Web Analytics implementation and focus on measuring trends rather than precise numbers. Remember that your web analytics can help you find out if your visitors are increasing or decreasing and with what rate they are increasing or decreasing.
Joydeep Deb is a Senior Digital Marketer and Project Manager with strong experience in Digital Marketing, Lead Generation, Online Brand Management, Marketing Campaigns, Project Management, Search Engine Optimization (SEO), Search Engine Marketing (SEM), PPC, eMail Marketing, Web Analytics, Web Technologies, Web Design and Development.
With an MBA in Marketing. IIM Calcutta Alumini. Lives in Bangalore, Karnataka - India.