"Average pageviews" is an expression used by webmasters to describe the average number of webpages viewed by visitors to a website.
It is one of the main website metrics offered by web statistics services such as Google Analytics.
The top source of income for many high-traffic websites today is display ads. The more ads are displayed, the greater the revenue.
Revenue is determined by total pageviews, the number of ads per page, and CPM rate (revenue per ad).
Total pageviews are the result of the total number of visitors and the average pageviews per visitor.
To increase total visitors, you get better rankings in search engines, and there is a lot of information online about how to do that. But what about increasing average pageviews? Most internet discussion focuses on rankings and finding the advertising networks which offer the best CPM rates.
Between rankings, CPM rates, and average pageviews, the statistic over which webmasters have the greatest control is average pageviews.
Rankings are controlled by the search engines. CPM rates are often fixed and non-negotiable. Average pageviews, however, are something you can change and develop because you have absolute control over your website.
Increasing the average pageviews per visitor is, in fact, the simplest way for display ad websites to increase revenue.
Webmasters of high-traffic sites should spend a significant amount of time (and possibly money) optimizing their site for maxium average pageviews.
One problem with high-traffic sites is small changes can result in large losses over the course of an entire day. For example, even a small percentage drop in pageviews could result in losses of hundreds or thousands of dollars. This high risk discourages change and even testing.
The risk of losses is greater when there's a statistics reporting delay. Google Analytics isn't very fast and it's not in real-time.
Based on the above assumptions and experience which exposed the reporting delay problem, it became clear the solution is a new statistics reporting script which is better than Google Analytics. It provides real-time average pageviews reports.
The key to useful sampling or short test runs is choosing the right sample size. A run that is too long can waste money if the site changes resulted in fewer pageviews; and a run that is too short can give you useless information which could lead to poor interpretation or missed opportunities.
How do you figure out a valid minimum or the ideal small sample size which truly represents the behavior changes on visitors, and at the same time reduces financial risk as much as possible?
One way is to find your baseline. What are your average pageviews to your site for one day?
Do several prelimary test runs without changes to your site simply to find out how quickly your site reaches a stable average.
First, you could do a test run for an hour. Then two hours. Then do intervals of 15 minutes; 30 minutes; and 45 minutes. If there is no difference in the average, you know that 15 minutes is a valid sample size (for that time of the day).
If you want to do a test run at another time of the day (e.g., early in the day or when traffic is lower), choose the number of visitors from the sample size for 15 minutes or whatever time matches the stable average point.
Testing on one site showed the stable average wasn't reached after 15 minutes; the average kept climbing for 30 minutes. That is when it started to plateau.
Instead of manually performing the above tests, the script can have a feature which automatically stops a test run after the site reaches a stable average.
The mathematical formula would calculate the percentage of change between two average pageview markers, e.g., Current Average (C) and Recent Average (R). If T/R X 100 > P (%), then the script stops. The Options from within the control panel allow the webmaster to set the Percentage.
Copyright © 2010 by Average Pageviews. All rights reserved.