Link Juice, which I shall use to refer to the combined weighting of link relevance, importance, and anchor text is a finite resource. A given web site has a finite amount of this asset, and then must decide how to allocate it. To refine this statement a little bit, you can think of a site as having internal link juice (from its own pages) and external link juice (from editorial links from other web sites).
With your internal link juice, you tell the search engines which pages on the site are the most important to you. There are two major ways to do this:
- Your information architecture (or internal link structure)
- PageRank sculpting (or siloing) usin NoFollow, NoIndex, and Robots.txt
But there are other aspects to this as well. Perhaps you have heard that one parameter to be aware of is how many clicks it is from the home page to your content. In general, this is good advice. However, if you make everything one click away from the home page, and you have a 100,000 page site, you have a lousy architecture.
For one thing, the search engine will only look at the first 100 to 300 or so links, so the rest of your site will never be crawled. In addition, it’s a lousy user experience. Last, but not least, this assumes that all your pages have equal value, and they don’t.
Ultimately, one thing you just can’t be lazy about in putting together a web site – deciding what the most important pages are. These are the pages that carry the most important messaging, or make you the most money. Figuring this out, and then designing an information architecture that tells both users and search engines that is best built in to a site design from the very beginning.
It’s a decision you must make, unless you want to let the search engines do it for you.