Wednesday, January 21, 2009

Supplementary Index Research Analysis

I've pursued more research and performed some original analysis on the concept of the Google Supplementary Index. In my last post, I had found how to get Google to tell me how many of pages were indexed versus how many were in the main index.

I then discovered how to find the pages that Google has put in the supplementary index. Go to your webmaster tools and for each page, look at the inbound onsite links and the inbound external links. If the site has some links but Google is NOT showing them (after a reasonable amount of time), this means that Google has decided that page belongs in the supplementary index which means:
  • It rarely shows up in search results
  • It's inbound links do not really contribute to your site total
  • It's internal and outbound links do not really count
  • And too many pages in this index seem to hurt your site's quality profile

So I've identified some pages on my site as being in this index and conducted some research about how to get them out. It's not so scientific but it is data. Here's the results.

Test case 1. Added 1 external link from a blog in the third position on a group of 27 links. Appeared with one link counted internally and one externally.

Test case2 . Added 1 external link from a blog in the 8th position on a group of 27. Also added it to my signature on digitalpoints.com forum on half a dozen posts. Did NOT get any links counted..

Test case 3. Added 1 external link from a blog in the 10th position on a group of 27 links. Appeared with one link counted internally and externally.

Test case 4. MD. Added 1 external link from a blog in the 12th position on a group of 27 links and rewrote page with very original text. Do NOT get any links counted..

Test case 5. MA. Added 1 external link from a blog in the 13th position on a group of 27 links. Used pingler to notify all it's clients about the page. Did not NOT get any links counted..

Test case 6. MI. Added 1 external link from a blog in the 14th position on a group of 27 links. Used pingomatic to notify all it's clients about the page. Did not NOT get any links counted..

Test case 7. NJ. Added 1 external link from a blog in the 15th position on a group of 27 links plus side bar links on other blogs that already link a lot to the target site. Appeared in both lists of links with 71 external links!

Test case 8. NY. Added 1 external link from a blog in the 16th position on a group of 27 links plus side bar links on another blog that already links a lot to the target site. Appeared in both lists of links with 105 external links!

Test case 9

. RI. Added 1 external link from a blog in the 19th position on a group of 27 links plus inside a few week old blog post on one of the blogs tha worked for 7 & 8. Did not NOT get any links counted.

Test case 10. SC Added 1 external link from a blog in the 20th position on a group of 27 links plus inside a few week old forum posts. Did not NOT get any links counted.

Test case 11. VT. Added 1 external link from a blog in the 20th position on a group of 27 links. Did not NOT get any links counted.

Test case 12. WV. Added 1 external link from a blog in the 21st position on a group of 27 links. Added a no follow link on a new blog. Showed up with two external links counted, one from the group above and one somewhat randomly from an unrelated site that is using algorithmic linking to many sites by topic and seems to have scooped up this page. Weird.

Conclusions:

On a list of links from a page, the first ones count more. This explains test case 1.
The side bars of blogs really count a lot. They worked everytime.
Old posts in blogs and forums work sometimes.
Changing content did not work.
The six links from my signature at digitalforums did not succeed.

No comments:

Post a Comment