When Web Analytics was still starting, the main data source of analysts which they use to analyze their sites with are from the website’s log files. They use a log file analyzer, an example of which is AWStats or Advanced Statistics, to which they derive metrics such as:

• Hits
• Page Views
• Visits
• Clicks
• Visits Duration
• Top Pages
• Bandwidth Usage
• File Types
• Visitors’ Browsers
• Errors

From these, they were able to measure a bit of the visitor’s activity but as this is a log file analyzer, web analysts benefit more in knowing website performance rather than visitor conversions. But pretty soon, more advanced Web Analytics tools surfaced to aid analysts seeking to get deeper into web data. The advent of these tools, such as Omniture SiteCatalyst and Google Analytics, brought in more defined and meaningful metrics. These newly-found metrics are of such high importance that web analysts started to disregard the metrics offered by log file analyzers such as those mentioned above.

They may have mattered before but web analysts knew that these only count numbers, the tip of the iceberg. These do not tell how much of these Hits and page Views were converted to buyers (in case of e-commerce sites) or were triggered to do a certain activity the sites aims for its visitors to accomplish. These conversions are some of the many metrics which advanced Web Analytics tools offered to produce and then web analysts have been empowered once more. They can now focus on what web analysts should be doing:

• Compute for different conversions
• Do Path Analysis
• Dive into A/B and Multi-Variate Testing
• Perform Keyword Analysis
• Compute for step-by-step purchase conversion
• Know the Average Order Value
• Check for site-wide conversion rate
• Understand visitor behavior and traffic trends
• Create other actionable reports

Here’s a perfect example of the process Web Analysts should take when doing their analysis, perfectly illustrated by none other than the Google Analytics Evangelist himself, Avinash Kaushik.

They can just then leave the log file analyzer to the development and programming teams who can really benefit from knowing bandwidth usage, visitors’ browsers, file types, and errors. This way, getting dirty with web data for the purpose of creating actionable analysis will really be worth the while of a web analyst.

One of the my main takeaways from this year’s eMetrics is that metrics is ridiculously political. Having the responsibility to put up a web analytics team for the past year is not easy. From going solo for the first few months, to expanding, to defining the role of the team, to being glorified report builders… And now, internalizing the value that web analytics bring to the table. The team’s identity was put on the line.

Data is very important to any company. It is where decisions are based. Of course it is only natural that ownership of that ‘official’ data will be a struggle. But once you realize that you don’t need to fight, you will be able to participate. Web Analytics is not a substitute to traditional data, it is a supplement, or a complement if you must.

Web Analytics is no longer just web data recorded in log files. It is an art of deciphering your customers, what they mean when they act a certain way, what they want when they do this and that. It is no longer what happened in my site, but why did it happen, and what can I do about it.

Experiencing politics in the field of Web Analytics is scary for me. But with the intent of bringing in good through the story we see in our numbers, then it becomes an adventure. And that adventure of ours here at WAP, is just beginning.