Arabian Business / Accenture Web Index - September 2001 Report

So what exactly is the best way to measure a Web site's traffic. Should you take into account page impressions, as the Arabian Business Top 10 has so far done? Or should you look more closely at other metrics, such as visitor numbers or the time that visitors spend online?

  • E-Mail
By  Peter Conmy Published  November 5, 2001

Page impressions versus visitor numbers|~||~||~|Click here to view the Arabian Business/Accenture Web Index for September 2001"

So who really is the most popular site?

So what exactly is the best way to measure a Web site's traffic. Should you take into account page impressions, as the Arabian Business Top 10 has so far done? Or should you look more closely at other metrics, such as visitor numbers or the time that visitors spend online?

The question arises following two major recent rises in page impressions from two of the Index's regular top three sites. Last month, Arabia.com jumped to 46 million impressions from around 37 million the previous month, following the addition of an Arabic language chat feature. This prompted complaints from some of its closest rivals, who questioned whether such a new feature could really boost traffic by that many impressions.
This month, it's Maktoob's turn to reach the stratosphere. Maktoob.com has risen from around 44 million page impressions for August to a staggering 60 million plus for September. This follows the introduction of news, ladies and cars sections in early September.

In both cases, Arabian Business has looked through the respective sites' reporting software and the statistics have tallied. So this seems like a good time to ask the question: Do rising numbers of page impressions mean that one site is more popular than another? Or could it just be differences in the reporting software that different sites use? Or maybe certain features, such as chat and forums, create more impressions in some reporting solutions than other features do.

Anyone making an online advertising decision looks at more than just page impressions, taking into account visitor numbers, target audience, geographical focus and time that users spend online on a site. A site where people spend a long time could be seen as 'sticky', and you could therefore argue that advertisers will be getting more exposure time for their money.

A site with lower numbers of impressions might have a higher number of individual visitors than its competition. In such a case, advertisers might look more at the number of individuals they're reaching than the amount of activity on the site.

Taking visitor numbers deeper, a site that boasts a total visitor number significantly higher than the unique visitor number obviously attracts people back. This implies a site that users are intensely loyal to. For example, compare Ajeeb with Arabia.com. Whilst Arabia boasts a noticeably higher number of impressions than Ajeeb, Ajeeb's number of total visits is significantly higher. Unfortunately, the comparison can't be followed through to its conclusion, because unique visitor numbers aren't available for either site.

There are so many ways of looking at how popular a site is. The three 'Top Fives' that we've published here show you how very different metrics produce very different results.

These tables are certainly not definitive, as many sites cannot supply visitor numbers and looking at the time spent on a site is something we've done more for fun than anything else. We feel that they do reinforce one thing, however: that a site's popularity is judged by many things and that advertisers have to think carefully where to invest their dollars online. In future, we'll be looking a lot more closely at visitor numbers.

Click here to view the Arabian Business/Accenture Web Index for September 2001"

Research by Annika Pohjola. Send feedback to: david.ingham@itp.net.

||**||Web Index: Glossary of terms|~||~||~|Ranking in category by page views: Web sites are ranked according to the total number of page views reported in the sample period.

URL: Internet address of the Web site.

Site description: Web site owners have submitted their own descriptions. These have been edited by Arabian Business.com due to space constraints.

Site category: Categories have been formed by Arabian Business.com for the purpose of comparing Web sites in a like-with-like environment.

Total number of visits in period: The cummulative total of visits to a site in the specified period. A person visiting more than once in the sample period will be counted every time that person visits. Visitors to a site for a prolonged period, i.e. opens a Web site and leaves that site open on his machine all day, can count as several visitor sessions in that day. Web tracking software can effectively call a visitor session closed if it remains inactive for a predefined period.

Average duration of visitor session in period: The time taken between entering a site and leaving a site.

No. of new visitors in period: A count of visitors who have not at any time been logged by the Web tracking software.

No. of unique visitors per month: Unique visitors are counted using the visitor's IP address, domain name, or cookie.

No. of page impressions in period: A count of hits to pages defined as documents or forms by the Web tracking software. In most, but not all cases, a cick to a new page within a Web site will count as a single additional page impression. The supporting graphics on pages are not counted.

Sample period: The dates during which the log data was generated.

Verified by: Note of whether the visitor data has been checked by Arabian Business.com and Accenture or whether the site owner has supplied data without independent verification.

Click here to view the Arabian Business/Accenture Web Index for September 2001"||**||

Add a Comment

Your display name This field is mandatory

Your e-mail address This field is mandatory (Your e-mail address won't be published)

Security code