"...The purpose in RankAttack technology is simple: get the search engines attention and make them want to list your website under the keywords you desire... "

It immediately peeked my curiousity enough to obtain a copy for evaluation. The software advertised at a price of $79.00 on the companies website with an equal price for its sister product SEO Sandbox (A product claiming to reduce a new site from the Google Sandbox). They offer a de-tuned trial version which allows full functionality with some limitations for the speed and duration of its operation which I promptly registered and downloaded.

With recent changes in indexing algorithims and apparant "quality filter" testing across the major search engines this software appeared to target a market desperate for relief. Claiming to have been developed by a software engineer at a "major portal" who was terminated for using the technique implemented in these products added to the taboo allure of the application. The method by which the software purportedly acheives improvements in SERPS is that it "...creates a persona of popularity..." around your site in the eyes of the search engines..." and maintains additional claims of baiting the spiders to your visit site by artificially increasing the effective query traffic for your keywords.

The obvious appeal of a product that can safely deliver the results advertised is too tempting to dismiss. However, the "safety" of the software should be of greater importance then the potential rewards if the cost incurred is that of search engine penalties or bans.

Without clearly defined technical specifications regarding the manner by which the software increases "...popularity..." for a given url and keywords I relegated to packet analysis of outbound and inbound traffic to try and determine if the product was in fact employing "white hat" or search engine safe . The results were sadly not surprising, revolutionary, technologically impressive, or safe in my opinion.

The initial setup is straightforward requiring the user to enter a url followed by a text box to input keywords. The subsequent screen offers a list of more black box options to add additional search engines, directories and 2nd tier engines to improve results. After maintaining the default selection of all options the application launches into a final screen with an option to limit how many threads to employ in addition to an option to "operate in safe mode" which throttles the initial threading prior to starting. The remainder of the process is entirely black box only provinding a report of the serach engines targeted and how many "queries" have been submitted.

The actual data sent is unfortunately equally simplistic. For testing I submitted the url of http://www.wpdfd.com which is currently ranked #1 in Google for the keywords "web design". The following is an excerpt of a tcp stream taken from ethereal:

GET /images?q=tbn:AjOIvuzQYsAJ:www.wpdfd.com/images/squares.gif HTTP/1.1
Accept: */*
Referer: http://a9.com/web%20design+web%20design+web%20design+wpdfd.com
Accept-Language: en-us
Accept-Encoding: gzip, deflate
User-Agent: Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; .NET CLR 1.1.4322)
Host: images-partners.google.com
Connection: Keep-Alive

The full packet log file can be downloaded in: Text (4.5M), XML Summary (980K), XML Full (39M)

The means by which the software appears to increase "...popularity..." is by submitting a massive amount of queries to each search engine in its database by taking the url and keywords provided. The algorithim used is apparently a trivial combinitoric permutation of submitting search queries in a cascading manner to each engine by spoofing the HTTP REFERRER from another engine in a fashion similar to spanning trees. This manner is of questionable ethics and even more questionable in results given the simplicity of the algorithim.

The first issue that come to mind is the potential for IP Address bans by the search engines for the machine executing the application as it is a clear abuse of TOS for all engines. Google has already employed IP bans on excessive queries without use of their API. Simply spoofing the referrer is not an effective method to mask the originating source address as perhaps routing through series of public proxies.

The claim of this drawing spiders to crawl your site is highly improbable even if run 2-3 times a week for 2-3 hour durations as suggested by the vendor. Furthermore the vendors site is highly indicitive of the caliber of "software" this is : http://www.savemoneyfast.com ; basically should warn a potential buyer.

Some warning signs potential buyers should be aware of:

* The site was established in Feb. 2005, registered by anonymous proxy and is currently without pagerank or backlinks within google. Highly odd for a site promoting SEO tools.
* The fit and finish of the website is sub-standard given the market segment being addressed.
* Use of Hotmail email addresses for support. Support forums were closed (however cached by google).
* No PageRank.
* No valid Backlinks.
* No posted company address, telephone numbers.
* Payments only via paypal to a nonrelated baby apparel address (intentionally ommited for privacy).
* Actively promoted on Ebay.com to unknowing an consumer base.

The software, vendor and results all leave more questions then answers. I would not personally use this software outside the scope of testing it herein. Nor can I see any possible benifit this product can have given the observed way in which it performs.

RankAttacks sister product SEO Sandbox was also tested with similar results. The manner by which this product "works" is highly similar however rather then just simply submitting many cascading queries its added complexity is that it submits additional "site:xxx.com related:xxxxx allinanchor:..." parameters in its queries and also has the "capability" to operate in hyper mode (basically increased throttle rate for threading. A more detailed review will be posted shortly.