Calendar
Network
Feeds
- 39
- Sphinn It!
Went Hot: April 4, 2008 - 1:21 pm
Posted By: theGypsy 457 days ago
Topic Type: News Story (Jump to http://www.blogstorm.co.uk)
Category: Link Building
* Ability to analyse up to 20 domains at once
* Tool can analyse 20 domains and find their most common link sources
* The tool should run in the background either via the web or a browser
* Show PR and link data for any page on a site
* Show PR and link data for any page that links to you
* Show anchor text and title tags for these pages
* Option to exclude nofollow links
* Graphs of how all link data changes over time
* Ability to filter links by domain extension
* Ability to filter links by hosting location
* Easy link to whois data for every domain
* Calculation of deep link percentage
* Use of colours to represent one way or reciprocal links
* Ability to exclude certain domains from the reports
* Ability to exclude sitewide links
6 Comments
Comments
Here's what I came up with (it's early still....zzzzz); from comments
hmmmm...lemme think.
Relevance; Ok, you have the link text and page title; for further relevance I would love a primary phrase count or other page theme info (like SEO Digger). Knowing the theme of the page is important IMO.
Page segmentation; I would also consider link location to take page segmentation into account ie; header link, footer, side panel or editorial link.
Temporal; Last (but not least...could go on) is Timing; since links degrade , it would be easier to identify valuations by knowing when the link was discovered, when the page was discovered/last updated. This speaks to the temporal value of the page (and link). Possibly also graph the link discovery (dates) ages to understand the freshness of the link profile as a whole.
..there's my 3 :0)
Page segmenation, thats a good one. And a hard one to do.
well hey, ya didn't say 'easy ideas' ;0) ... just some things that came to mind first thing in the morning he he.... I shall keep thinking as the day goes along.
The biggie to me would be a check via archive.org or similar on the age of domain. Search history plays such a crucial role in link authority IMO, and archive records - excluding SEDO parking pages and similar - would be a good help on approximating that.
There is absolutely no better tool than the human brain and good pair of eyes, though - link building reduced to number crunching easily creates fundamental problems of not understanding the context of a site's ranking.
2c.
I'd love to see it as a Nth link on the page or Distance(in pixels?) from the top. Those are the obvious drivers. If you cna extract CSS information (size, etc) also that would be cool, if not hard.
Maybe some sort of automatic trigger to spot link farms?