I love big sites. They are like sprawling mansions in a state of renewal and dilapidation at the same time. New kitchen, old bath. Sub-zero inside, rotten wood outside.
The longer a site has been around, and the bigger that site is, the more likely it is that site will have already attracted hundreds if not thousands of links. Many of these existing links will have come about without anyone actively pursuing them. For example, I’ll bet Coca Cola’s web site has a section for employment, and I’ll bet that employment section and the pages within it are deep linked to by other sites. I’ll further bet that Coca Cola likely does not know how many other sites are linking to their employment section pages, nor have they checked.
Many large scale corporate sites have multiple content sections (AKA silos) with different personnel responsible for them. Over the years, these sites often go through several staff changes, teams, agencies, etc., and end up with an enormous amount of inbound link equity.
One of the unfortunate facts of life for large sites with many bosses is sometimes content decisions are made that have a negative impact but nobody knows it. Content gets neglected, even forgotten. The classic example is the page or section that gets killed off without replacement, when no 301 redirect is used.
Without a redirect, the server ends up reporting a 404 page not found error to the user, which is duly logged. But the frustrated visitor is the only human likely to notice the problem, unless the company has assigned someone internally to monitor for such requests. Worse, monitoring can be difficult when a site is serving millions of pages a week and has server logs so large they are purged on a weekly basis. Sound familiar?
I understand part of this problem is inevitable and the cost of doing business. But I try to educate that sometimes the answer is not to just go after more links but rather to develop a sensible link reclamation strategy, something I first spoke about back in 2005 at an SES conference. Not everyone needs educating, however. More and more, as the lost link equity problem has grown for large sites, I get specific request for this type of help.
One of the more useful things you can do to chart a course towards link reclamation is start first with those links that produce 404 errors on your site (both Google Webmaster Central and Microsoft Webmaster Center offer tools that can help). Even a 404 page request can have a referring URL. Sort the errors so that those happening the most are listed first. Then track down referring URLs to these pages. You might be pleasantly surprised to find, like I recently did, that those links might be located at extremely trustworthy and powerful places.
That’s some extremely useful data to have from a link reclamation standpoint. It will also make it easier for you to justify and argue for the page to be recreated or at the least redirected to other pages.
By killing a page with trusted link equity, you killed a bit of potential and real rank along the way. Over time, and staff changes, and content changes, that can really add up. Like holes in your driveway. Or 404s.