Format of a cloaked page?

[Ash]

Hi, I,ve read the cloaking faq thats available on this forum, which is very good. It's good to be able to read the plain facts on cloaking without the usual hysteria that accompanies them. I have also read quite a bit from other places on the web but I've never actually seen anyhing concrete about the format a cloaked page should take. For instance are H1 tags better on a cloaked page than the H4 tags I usually use, if so why? I've read that a cloaked page should be no different to a doorway/marketing page, does this mean no different whatsoever, just no images and graphics or what? Thanks for any help.

[Air]

IMO the cloaked page should be a stripped down version (html, css and javascript wise) of the page that is being shown to regular non search engine visitors. In addition all graphical titles and headings should be replaced by their text equivalent.

The other opportunity that a cloaked page affords is to introduce some elements of html like bold text, italics, underlined text, etc, that give a search engine some idea of what is important on the page. These may look ugly or inappropriate on the regular visitor page and would not be used.

As for the H1 vs. H4, there is some real and anecdotal evidence that H4 is just as good as H1 as long as no H1,H2,H3, tags appear on the page. In other words the algos appear to give weight to headings within the page rather than across sites. But since only search engines will see the ugly text generated by an H1 tag on a cloaked page, it seems using H1 for titles and maybe H2's for headings (and maybe an H3 around a keyword phrase ;) )will produce the desired result.

<added>Welcome to WebmasterWorld! Hope you enjoy the forums.</added>

[Ash]

Absolutely spot on Air, that was just what I was looking for. Thanks very much.

[Filipe]

Enough with the anecdotal evidence! Has anyone run any real tests on these things?

[Filipe]

Heh, that are willing to share that is.

[volatilegx]

Hehe not me! :) If I had a template that worked every time (not saying whether I do), It would be worth it's weight in gold! Wait a minute... templates don't weigh anything... OK worth my weight in gold!

[Ash]

Yeah I ran three test pages for a fictional client using some spare url's with the information Air posted. They are ranked 2, 4 and 17 atm on UK se with 85000, 15000 and 180,000 competing pages respectively. Obviously these rankings could change but the same keywords haven't achieved these results with marketing pages that I've optimised for a real client.I'll keep you posted on the rankings.

[sinyala1]

Anyone working or have seen any spider algorithm cracks? Or anything that comes close to it? I'm in the midst of programming one and almost ready for testing. Has anyone really used, seen, heard even a whisper of such a thing? I need some one to run back and forth off ideas and things that have worked and haven't.

[Filipe]

Spider algorithm cracks? Either you mean

1. A spider simulator or
2. Working data on a spider's algorithm

There are a couple of the first, practically none of the second. If there were hard data on spiders' algorithms, we wouldn't have to discuss it on this board, and the web would be a mess :)

[jeremy goodrich]

sinyala1 - I think we've talked about such things here before.

To determine the format of a cloaked page that you are going to create, one of the best ways is to run a 'cracker' over the SERP's of the engine you are interested in ranking high on.

To do that, you put together a list of variables you are going to parse out of the analysis of the pages in the ranks that they have achieved on the engine you are targetting....then you find the similarities across the pages. Once you've got that down, you can spot 'trends and patterns' on the charts that you put together.

From reading some of your posts, you sound a fairly knowledgable programmer :) it should be pretty easy to figure it out from there. Try a site search on 'cracking search engine algo' or perhaps 'brett tabke's eggshell (or egg shell) that's a quote from a post I made in a thread discussing similar.