Who Should Use This Document

This document provides a comparison of SmarterTools’ SmarterStats Web log analytics and SEO software and the Google Analytics website statistics service. The comparison is from the end user’s point of view.

Overview

Regardless of whether a website is designed for a business or individual, it is important to collect and evaluate the site’s analytics. At a very basic level, this involves tracking information about unique visits, page views, bounce rates and conversion rates. Analyzing this data will help you predict visitor behavior and modify your website design and/or marketing campaigns for improved success.

Methods for Gathering Website Analytics

The Web Analytics Association defines Web analytics as “the measurement, collection, analysis and reporting of Internet data for the purposes of understanding and optimizing Web usage.” There are two main methods for collecting website analytics: log file analysis and script-based analysis.

Every time someone visits a website, the Web server records information about every file request, i.e., the HTML files, CSS files, JavaScript files, graphic files, PDF documents,MP3s, etc., and stores them in log files. These log files also include information on site and Web server errors, page processing time, bandwidth used, visitor IP address, referring websites, and much more. Log file analyzers like SmarterStats interpret this data to create website statistics and reports.

Alternatively, script-based analyzers like Google Analytics require the use of tracking code (usually JavaScript) on each webpage that is to be tracked. As visitors surf the website, the code places a cookie on their computer so they have a unique identifier and can be tracked—provided the visitor is using a browser that supports JavaScript, has cookies enabled and is not using security software to protect the browser.

Log Analyzers vs. Script-based Analyzers

Each method for gathering website analytics has its benefits and drawbacks. For example, script-based analyzers only record data from pages that have the JavaScript tracking code; any pages without the code will not have statistics available. This is also one of the reasons log file analyzers report higher traffic (views, visits and hits) than Google Analytics does, as the Web server log files track every interaction visitors have with a website.

Similarly, because log file analyzers process log file data instead of relying on forward-facing JavaScript, they have access to information that script-based analyzers don’t (such as data relating to traffic from spiders and bots, server errors and bandwidth statistics).

Accuracy of Statistics from SmarterStats and Google Analytics

As previously discussed, users of both analytics program may notice reporting discrepancies, particularly in regards to visitor counts. Because of the limitations of script-based analytics, Google Analytics will underreport or misrepresent some website statistics. Consider:

• When comparing data on website visits, it is important to realize that SmarterStats and Google Analytics process data at different intervals. Google Analytics uses a 30-minute window while SmarterStats uses a 20-minute idle timeout window. For this reason, SmarterStats will report a higher number of visits for visitors that keep the browser window open but do not navigate to new pages within the window.

• Look at the referral reports in SmarterStats. Are there any referrals listed that are part of the base domain (i.e., IP addresses or aliased domains)? These referrals are not available in Google Analytics reporting.

• Hits from people “borrowing” content from your website. Are people linking directly to documents or images on your website? SmarterStats counts any request for those files as visits and hits; Google never reports them because JavaScript can’t be placed on files, just HTML pages.

In the end, SmarterStats provides more accurate website statistics than Google Analytics because SmarterStats reports all requests to the server, not just what is tagged with tracking code.