TheMadScientist - 11:06 pm on Dec 13, 2012 (gmt 0)
Alright, well I've noticed some interesting discussions in a couple of relatively 'general advice' threads that strayed a bit OT, so after a bit of discussing with some of the members involved I thought we should have a dedicated discussion about what I'm noticing...
The first are some 'key points' from the 'Sandbox Effect' thread:
These domains are ranking exclusively on their backlinks (and rapid acquisition thereof).
In the past, it was common to see brand new domains soar to the top rapidly, then move down in rankings; but, I've watched some of these domains rank well for several months now.
It almost sounds like we're seeing a 'shift' to visitor behavior having a greater influence than it has previously based on what you're saying.
Another interesting characteristic of a few of the sites I am talking about is that some lack useful site navigation ... There no links to internal pages (except for a "contact" link). The domain was registered 4 months ago. Furthermore, the internal pages contain no links to each other, only a link back to the homepage...
I don't understand what it is about the site that makes it "useful" to a visitor (or what metric could define it as such), unless an immediate click to an affiliate link equals visitor satisfied (or if the visitor returns back from the affiliate site only to click on the next affiliate link in the list, and so on).
But in the case you're talking about (immediate click to an affiliate link), to Google, the visitor 'disappeared' and did not return to the results, which would almost have to be interpreted as 'visitor satisfied' by an algo, even if it doesn't make complete sense to us WRT 'the site clicked' satisfying the searcher, because we know it really didn't.
how might the algorithm might interpret affiliate links opening to target="blank" when it comes to visitor behavior?
It can't, because it doesn't know ... You might know, I might know, but the algo knows 'when the visitor returned to the results', 'what did the visitor search for upon returning' and 'did the visitor block the site upon returning', that's it, nothing else.
I have the same observation. I am looking after a site that sells circle and diamond widgets only, in certain geographical area. The site has decided to create a page on square widgets, reviewing the square widgets in the same geographical area. The page contained 15 different square widgets with photo, main features and dofollow link to the square widget manufacturer (there is only one of these for each square widget)...
After creating the page, it initially ranked at the bottom of the second page for the keyword square widgets geo-area. Then over the course of the next 6 months it started to climb, ending at #1 and holds this position for the last 2 years. The bounce rate of this page is 70%+
And from the recent '404 / Broken Link' thread:
Just did a test of this on a #*$!ed website. When you serve a blank page with 200 OK header google drops the page from the keyword index entirely in my test...
I was just experimenting, I thought the keyword would still rank on a blank page due to inbound links. But I was wrong
What I 'get' out of these two discussions is:
Either Google is missing the 'spammy backlinks' crobb305 and a couple of us who have seen the sites in question noted, which should, in my opinion, be easily detected algorithmically, because I'm fairly certain I could code it, so thinking the programmers at Google have not thought of and figured out how to detect those type of links makes my head hurt a bit...
Or (when I put it together with the test posted by seoskunk in the 404 / Broken Link thread is: It seems to be a fairly definite indication page content (or lack of content) can override inbound link text, which as tedster noted could be to help prevent Google bombing, but could also indicate there's a bit of a shift in importance for rankings from 'links as the way to go', to other on page and visitor behavior metrics having enough weight to override inbound links, even those that may be determined algorithmically to be 'questionable' or spammy.
So, I think the question I'd like to 'start things off with' is:
If you look at pages ranking well in whatever niche you're in that don't appear like they should be where they are and you forget about the links (just ignore them and look at anything else you can think of), what other things do you see you think could cause the page to rank where it is?
To me personally it looks like 'link value' can definitely be overridden by other factors, so I think it would be great to discuss what other factors people think could be in play if we 'throw out the links' to a page.