Website Traffic Measurement

There are two recognised ways that website traffic is currently measured - site-centric and user-centric measurement.

Site-centric measurement

Here, statistics are derived from the records of activity captured by the site. Typically, these records are the log file records of the site's host server(s). Site centric delivers a census approach: measurement of total activity on a site for a given period. This approach enables a site to give an independent third party auditor access to that data, such that they can then verify:

a) the authenticity of the original data records, and
b) the accuracy of the statistics derived.

This can deliver both pre-buy (web server audit) and post buy (Ad server audit) statistics. Sites may use page tagging (see below) to generate the log file records that then act as the source data for the audit. This is sometimes referred to as browser-based measurement, since the log record for the page tag is only written if a browser successfully requests it. ABCe (www.abce.org.uk) conducts audits on both site server and tag server data.

User-centric measurement

Here, statistics are derived by extrapolation across a defined universe of activity recorded by a panel of internet users. This is typically based upon data collected electronically through a device attached to the panel member's computer (PC) that records the sites visited by the User. The panel is built to be representative of the defined universe; its size is set to deliver statistically significant results.

Such research is currently conducted by a number of commercial research companies in the UK, including Nielsen NetRatings (www.nielsen-netratings.com) and ComScore (www.comscore.com).

Another approach, sometimes known as network-centric measurement, seeks to monitor site usage by sampling at the ISP level. This is adopted by such companies as Hitwise (www.hitwise.co.uk).

These approaches are, by their nature, not census-based and hence not auditable by ABCe. They represent different types of market research.

Auditing vs. Research - which one should I use?

Both approaches have a valid role to play in helping to generate greater understanding of the activities of Internet Users. In traditional media both approaches co-exist, for example circulation audits and readership research. Both offer a way to make comparisons between competing websites.

The primary difference is that audits offer sites the ability to have an independent third party verify their statistics, based on the actual total activity recorded by the site. Research, by comparison, delivers an overall picture of User activity, typically broken down by site, derived and reported by the research company. Results in both cases depend on the methodology employed. For more information on Research refer to the companies involved (above).

Auditing

ABCe (www.abce.org.uk) is the UK and Ireland industry-owned, not-for-profit organisation which carries out audits across all forms of electronic media. ABCe uses globally agreed Industry standards (see www.ifabc.org) as the basis for site certification. Only sites that have achieved a successful audit are certified. Auditing is voluntary, though sites that register with ABCe are encouraged to achieve 2 audits in a single year, the minimum frequency recommended by JICWEBS. Currently all audits remain confidential unless the site wishes to place the information in the public domain. All published audit data is available on the ABCe website at www.abce.org.uk .

Certification by ABCe delivers statistics to global industry-agreed definitions for metrics including Unique Browsers, Page Impressions, Ad Impressions, Ad Clicks, Searches and Visits. ABCe can also certify Page Impression totals for various content- and device-types. These currently include Bluetooth, WAP, Interactive TV, Flash, automated content, media streams, and chat. Unique Browser totals for some of these platforms can also be certified.

Other Metrics
As well as the above metrics, audits can deliver geographical and demographic information where the site captures such and is able to deliver an effective audit trail for verification purposes. Audits can also deliver information about the types of device making requests for content, and about the type of content delivered (such as mobile).

For more information contact info@jicwebs.org.