Comments on: Are you misusing Alexa numbers? (Probably)

By: ZK@Web Marketing Blog

ZK@Web Marketing Blog — Sun, 03 May 2009 22:23:40 +0000

Alexa is not that good, the best option available to configure the quality of the site is seoquake. It give some of the most exact and accurate results which are responsible to the fine optimization of the site.
Alexa has received critics from around the Internet for a long time already, but lately it appears that such criticism is intensifying.
Tech Crunch recently published an article titled “Alexa says Youtube is now bigger than Google.” The conclusion from the article is straight forward: Alexa is useless.

By: ZK@Web Marketing Blog

ZK@Web Marketing Blog — Sun, 03 May 2009 15:23:40 +0000

By: Scott Germaise

Scott Germaise — Sat, 05 Apr 2008 20:28:41 +0000

There are at least a couple of other issues I don't think you've mentioned.

* Outright Fraud *
There are organizations, (business or otherwise), who fully understand how the toolbars work. They can unscrupulously have hundreds or more folks from wherever download and use these toolbars. The companies may be aware of certain statistical anomalies from individual or blocks of IP addresses and blacklist them. Still, some of the black hat crowd are fairly sophisticated about this. This may be foolish and in some cases - with regards to mis-representation - possibly even illegal. (Or at least civilly actionable if you use artificially inflated data to make claims resulting in financial gain; or loss, as the case may be.)

* Small or New Sites *
If using data from these sources for business planning and competitive research, great. (As long as you take into account the caveats.) But for small or new sites, none of these methods is all that great. The sample size is either too small does not have nearly enough representation of niche segments. Anyone who's worked on small or even medium sized sites, (even in the millions of unduplicated users and millions of page views), can see this clearly enough by comparing their own stats to even rough extrapolations of the data from these services.

You made some points about how to use the data. And I fully agree with those. As long as your honest - with yourself as well as others - as to the huge hunk of salt with which you take the data and how you use it, that's ok. As there's really not much else to go on.

Scott

By: Shanti Braford

Shanti Braford — Sun, 06 May 2007 18:00:58 +0000

Just a heads up: if you have the Firefox SearchStatus plugin installed, it reports data back to Alexa.

Most Firefox users would not have this installed. I couldn't live without it though! It shows you every site's Pagerank & Alexa ranking, with quick links to lookup the site on Alexa, etc.

By: Kulveer Taggar

Kulveer Taggar — Sun, 06 May 2007 05:36:14 +0000

I spoke to one of these companies on the phone and they told me that they pay the ISPs (e.g. AOL) for anonymous logs of internet traffic and analyse that, as part of their overall way to work out traffic. You don't mention that anywhere?

I found it slightly worrying, but at least it's likely to be accurate.

By: MarketingBlurb

MarketingBlurb — Thu, 22 Feb 2007 22:06:55 +0000

PodcastPickle vs. PodShow

Earlier, Paul Colligan posted about how PodcastPickle recently passed Podshow in terms of reach according to Alexa stats (which have some problems but stay with me for now). The key difference between the companies as Paul points out is that…

By: djchuang

djchuang — Wed, 08 Nov 2006 01:49:23 +0000

But the problem is that web traffic measurement is a rather technical matter, and the decision-making executives often have their eyes glaze over if they can’t have a simplistic number to measure web traffic. Hitwise.com has what seems to be a better alternative to measure web traffic, by taking measurements from an ISP rather than from a statistical sampling.

And Alexa.com themselves disclaim the meaning of their traffic ranking for those rankings over 100,000 — which are often those websites that can’t afford to pay the $40k for the hard data to get a more accurate reading of web traffic and popularity:

“Traffic Rankings of 100,000+ should be regarded as not reliable because the amount of data we receive is not statistically significant.” (excerpted from http://www.alexa.com/site/help/traffic_learn_more)