Eric Enge has an excellent interview with Matt Cutts he just recently posted. The interview is about medium length and discusses some timely topics in depth. As usual, Matt is very forthcoming in his answers and some of the information provided is extremely valuable.
Here are what I consider to be some of the highlight points from the interview in note form:
- The number of pages crawled by Google is proportional to the Google PageRank of the website and its pages.
- PageRank decays as it moves deeper into pages.
- Minimize shared bandwidth to avoid failed crawled attempts due to host load limits.
- Duplicate content is merged into a single unit, not dropped or ignored completely.
- Avoid using Session IDs when possible.
- There is some PageRank loss in using a 301 redirect.
- The rel=canonical tag is okay to use with frequency and recommended when no 301 redirect is possible
- A blocked URL in a robots.txt is still indexed by Google
- Cutts recommends just letting Google crawl a website and sift out the duplicate or extraneous content. Excessive use of the robots.txt is unnecessary.
- PageRank sculpting is not necessary if a site’s architecture is sound.
- PageRank sculpting is being reduced in importance because Google wants search engines and users to see the same website linkage and have a similar experience when navigating the website.
- Be sure to use NoFollow links for ads on your website. Google does not want website advertisements effecting search engine rankings.
- Google has new link spam tools and technology that they will be implementing.
Of course, much of the information above is already known by most SEO professionals or was long suspected to be the case, but it’s always good to get an official word from Google itself on these matters.
