For adult website operators concerned with acquiring free, organic search engine traffic, having related sites linking to yours is an important factor in determining your website's relevance — and thus, how it ranks within the search engine listings.
While tools such as Google Analytics and good old-fashioned server logs will show you the sites and pages that are actively sending you traffic over a given period, many sites might have inbound links that are not very active — in other words, pages that link in, but that do not send any significant amounts of traffic — or perhaps even none at all.
I really might be stating the obvious, but it's something that I haven't read about before and in fact only considered yesterday: Some search engines may only count those links that are actually sending traffic to your site when determining the number of inbound links with which your site is credited.
At least that's my theory, and I'm going to tell you why: There's a tag that you can enter into some search engines that will reveal which sites that particular search engine "sees" as being linked to your site. It's the "link:" tag, and to use it, you simply enter "link:yourdomainname.com" (without quotation marks) into the SE's search box — simple enough, and something I've done before.
But when I ran two of my domains through yesterday, I had a bit of a surprise; on the first domain, Google reported that I had two inbound links. Yep, two — but I know for a fact that isn't the case. Indeed, when running the same command on the same domain name at Yahoo!, I was presented with a list of 442 inbound links — a huge disparity.
For comparison, I ran the second domain and received a whopping seven inbound links on Google, while Yahoo reported 18,493 inbound links.
Why is there such a huge difference between these numbers? I've been searching for an answer, but haven't found one yet — and the best guess I can come up with is that Google isn't counting "unproductive" inbound links.
Then again, Google doesn't seem to be counting some of my productive in-links, either, as some of the referring domains that I see in my server logs are not represented when I run the "link:" command at Google.
It's a mystery, and I don't get it. But who cares? Well, I do and you might, too...
You see, these two sites were built in their current incarnation for a specific purpose: to be organic spider traps. No "black hat" SEO tricks here — the strategy was simple and to me, more than legitimate: If the search engines rank sites based in part on the number of other sites that link to it, and the way they link to it, then I'll give them what they want, by encouraging countless relevant sites to swap links with mine.
The solution I came up with was an open-submission, text-link-only niche TGP/MGP that required a text-based reciprocal link.
By leaving submissions open to anyone, I wasn't limiting the volume of submissions, and thus increasing the number of reciprocal links pointing to my site. By requiring text links for gallery listings, I had pages of keyword-rich content that the search engines could easily spider and as an added bonus, eliminated 2257 concerns, as well.
Rather than using a graphic for my reciprocal link, I required a text link, which had my keyword in both the site title (and thus link text) and the URL — "pluses" from the SE's point of view, and a simple strategy that should have had beneficial results.
While I had hoped to receive some traffic from gallery surfers hitting my reciprocal link and then visiting my site, my real hope was that tens of thousands of inbound links with my targeted keywords in them would dramatically help my rankings for those terms and give me a steady stream of useful, high-quality search engine traffic.
This strategy was all based on supposed "common knowledge" of how the search engines may give one site more relevance (and thus better rankings) than another. Indeed, most of the traffic to my sites comes from organic SE listings, but my results could still be better.
Improving these results is something I'm now working on now, which led me to running the "link" command — and got me where I am today.
The upshot of these recent revelations is that regardless of how many webmaster forums and search engine marketing blogs you read or the information you glean from them, the theory and practice of SEO are two very different things.
I'm doing some pretty in-depth research and running some dynamic testing in an effort to refine my search engine strategy, and I'm no longer taking someone else's word for what is going to be a productive approach to the engines, circa 2008.
One of the things I considered is the relationship of these reciprocal links; sure, my TGP script checks to be sure that my link is actually on the submitted gallery and in the format I specified, but it doesn't look at the other reciprocal links on that same gallery and where they may point. This is important, because the way these reciprocal links should work is that one submitted gallery displays reciprocal links to three or more other TGPs/MGPs that the particular gallery was submitted to. This way, several TGPs/MGPs can share traffic, and thus site operators and gallery submitters can all benefit.
Unfortunately, some gallery submitters place your reciprocal link along with several other "recips" that point to their own TGPs/MGPs or other hub sites — but these galleries do not actually appear on those sites, so your site ends up sending traffic to not only the gallery, but to other competitive sites that simply aren't returning the favor.
Could these "orphan" pages, not having been visited by spiders before, have their links discounted? It was my first thought, but Google does spider my site and would have then followed the links to the galleries and seen my recip. After all, Yahoo! did.
Was it that these gallery pages, only linked to by my site and no other, would have their reciprocal link discounted? If that is the case, then the percentage of submitters that are cheating the process is astonishing, indeed.
At the time of this writing, I'm still looking at all of the data and doing some testing: If the level of submission cheating is as high as this cursory examination may indicate, and if there really is little to no benefit to this approach, then closing submissions could be the best course of action. At the same time, if Yahoo! is giving me some love over the recips, I won't drop them just because Google is ignoring them.
At the end of the day, I have many new questions about reciprocal linking relationships and their impact on search engine rankings. If this is something that you haven't fully considered, then you may want to run your own tests to determine if the traffic you're leaking through these links is justified by an equivalent or superior return. After all, if you give a little, you should be getting a little, too.