I've started thinking about link circuits more than link pyramids. Before defining the former let's look at the latter (link pyramids) which most folks are probably familiar with. But for the uninitiated let's borrow Jonathan Leger's description:
Google's algorithm is all about ranking sites that are "naturally" popular. That means that, although you can sum up how to rank your site in Google in four words ("Get lots of links"), you need to make sure that your site's link structure is as "natural" as possible (in Google's eyes, anyway).
To do this, you want to construct what I call a "link pyramid" for your site. This pyramid is a 3-level structure of links: 1) your base links, 2) your mid-range links and 3) your highest quality links.
Note my own schematic of the link pyramid in figure 1 of this post. Apologies for the lack of labels, Inkscape was choking up on me from all the detailed little page icons piling up. But you get it: the top of the pyramid, the yellow pages, are our highest quality links followed by some blues ones (mid-range) and the last pile of shit is our base links.
If Rand's indexation cap theory turns out to be true, the per-domain link pyramid is mostly a useless tool. I say this because when we think of per-domain, the amount interesting and unique content is bound to be very finite in Google's eyes. Content of a trusted distributor is bound to branch into different brands or images and thus different domains housing different linked to pieces of content Your scraped articles are going to fall out of the index, as well as that once cool page had some short-lived viral traffic from a stumbleupon PPS (pay per stumble?) ad campaign a few years ago. As your tree amasses content, expect the gardeners keen on original, unique and concurrently linkable and trusted content for the masses (I really only hope that's the over-arching agenda of the SE giants) to trim back the leaves of your content tree (in this instance being a site).
In a sense, the link pyramid is a incestuous web ring only imploded into a single. On the outset it appears completely white hat. You buy an old domain. You note the demographics. You ride on domain age of the domain and then begin building higher traffic to you best converting pages aimed at your accidental demographic. (There are still truly shitty pages get position 1 in the SERs by dent of a good domain name and site aging.) You adjust the link structure to funnel juice to your next pages hoping to yield good conversion. Rinse and repeat.
I think this is becoming less of a viable options as we head into what appears to be a new era of search engine algorithms. Google must well already understand the proliferation of wholesale content development and ballooning sites with scraped and remashed content or stuff fed out to Mechanical Turk and the likes. Perhaps that will truly be just months ahead of us in a
new era of information spam.
My own solution to making sites recognized by search engines in the new era is to think about remarkability, uniquess at a non-machine generated level, and diversity of content across many domains. I think it's also important to think about the form of the content itself. Text content will remain important but I think we need to consider all forms of multimedia and how that will play into indexing in days coming. HTML5 introduces native audio and video to browser and metadata of these will become crucial for Google to link the multimedia elements into its indexing and A/V verticals. I'm not alone in my belief that the web user (even the typical Google user) is shifting away from the reader or researcher to the more common distracted (ADHD sufferer perhaps). The Internet is becoming TV. We want a quick Facebook update, an flash game, some songs to listen to while we IM a friend, or a funny video, ...
Oh and what about link cicuits and moles? I'll go over this idea more in a future post. The idea is simply a platform for experimenting with new media and domains for distribution of media and testing indexability of new content. It plays not all into direct link-building, but ... Err, more details next time, but take it all with a grain of salt; really just a gut theory.
