Showing posts with label metrics. Show all posts
Showing posts with label metrics. Show all posts

Saturday, October 11, 2008

Competing for better web metrics

We have posted before (the last time just over a year ago) on the inadequacies of available web metrics for B2B sites. We publish a regular report as part of the BSG Asia Business Media Tracker subscription service which is based mainly on Alexa data. Some of readers and clients don't tire from pointing out to us the inadequacies of that.

So, I was interested this morning to come across a piece on Australia's TechNation site which gives traffic numbers for Aussie start-up sites. That's interesting in itself, but what caught my eye was that are using both Alexa numbers and those from compete.com. Call me ignorant if you will, but I wasn't aware of this one. I ran a test on numbers for Alibaba.com, Global Sources and made-in-china.com. I've always been a bit suspicious of the Alexa numbers for the latter. See what you make of the result:




I haven't looked into this seriously yet to work out where they get their numbers from. Have any of you? What comments to you have on Compete?

Monday, October 01, 2007

B2B web sites - where to look for 'real' data

Regular readers will know that we've written on the topic of the inadequacies of our reliance on Alexa.com ratings (most recently here, back in April). The main problem we have is that really don't see any better alternatives.

The China Web 2.0 Review blog has an interesting post today on this topic. It talks about the potential to use search engines to compare web sites' popularity. The basic principle appears to be that the more times people have searched for a particular site would indicate (a) that it is more popular than others and (b) that its traffic may be comparably higher. Those who rank poorly on these tests would, of course, argue that it's simply because those doing the searches are not regular users and that may be a fair point.

The article refers to Baidu Index, Google Trends and Yahoo Buzz. Play around with them. I got some interesting results which not be entirely popular everywhere.