Are you a Google Analytics enthusiast?
More SEO Content
Posted 13 September 2010 - 11:26 AM
Posted 13 September 2010 - 12:12 PM
Of course, if you're asking about links that you requested and you really need to know when they were created, just go to the page you requested the link from every day and check to see if it's there. But again, this information isn't really going to help you much.
Posted 13 September 2010 - 12:26 PM
Posted 13 September 2010 - 12:45 PM
The page linking to yours may have changed in other ways -- it may have removed other links, added other links, gotten links itself, or changed in other ways.
Pages you're competing with for that rank may have changed.
The algorithm itself may have been tweaked in some way.
Posted 13 September 2010 - 12:57 PM
Posted 13 September 2010 - 05:18 PM
I think you mean "credible". I have to disagree with you. I think you can test theories and rankings in an uncontrolled environment, but you have to allow for the fact you're not isolating parameters. In other words, you cannot draw statistical correlations between what you see and what you know. The data is simply too dirty or noisy for that kind of analysis.
But you can study trends. It's very similar to gravitational lensing, which uses the fact that galactic-level gravity forces distort light in certain ways. By accounting for those forces in their observations, astronomers are able to extend the power of their telescopes to observe distant parts of the universe through inferential data.
If you want to study links, you need to step back and assume that links are irrelevant to anything. That sounds counter-intuitive but you have to make your data prove to you that links are indeed relevant. You can't really do that if you begin with the assumption that links have some sort of impact on anything.
So instead of counting raw numbers of links (which tells you nothing useful or interesting) or even link inception dates, you might want to capture snapshots of link profiles every couple of weeks and chart trends for about six months. Don't make any inferences or draw any conclusions during that time. Just collect the data.
At the same time, you can take random samples of links you know exist and test them to see whether they show up in search engines, how many ways they can be found in search engines, and chart their visibility in search engines over a timeline (six months is fine).
In this way, you'll end up with a lot of data. But don't make the mistake of using Yahoo! to determine what Google knows about a link. Yahoo! can only tell you if Yahoo! has found a link. Bing can only tell you if Bing has found a link. Google can only tell you if Google has found a link. You cannot use any search engine as a reliable tool for evaluating what some other search engine knows about or is doing with links.
You also need to learn how to use advanced query operators and what they mean. For example, inanchor reports only listings that have links pointing to them (in the search engine's opinion) that use a specific anchor text. Many people in the SEO community wrongly assume that inanchor means something else.
You want to learn how to use intext/inbody, info, inanchor, intitle, inurl, the requirement operator (aka the "plus" symbol) and the exclusion operator (aka the "minus" symbol or hyphen), and you want to learn how to use quotes, site, and pretty much every other query operator.
You won't get complete data from any of them. You WILL get some confusing data.
The more correlations you can show between results from different query operators, however, the better. Not correlations as in statistical measurements that determine probable relationships -- rather, similarities in results.
Knowing that a linking resource appears for eight different types of queries tells you something -- something different from knowing that it only appears in two types of queries.
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users