I don't understand how the API count works. Why is there such a huge difference between the pages a normal site: command has and the ones the API has? Why does the coop use the API count and not the search: count? I was just about to buy a huge site with over 3,000 indexed pages to use for the coop, but after doing a check with the API it shows only 45 pages in url.
Beats me... ask Google. As far as why the co-op uses the API vs. www.google.com, it's because using www.google.com to make automated queries violates Google's TOS.
My theory is that the API returns a count of the pages that are fully indexed. google.com returns a count of the urls that google knows about even if they are not fully indexed. Notice on large sites that many of the results for a site: command are just urls that are not indexed. I always try to look at the API results if I can. Ross
Many times the API predicts what will be happening elsewhere. A drop like that would usually tell me that they are about to loose a ton of pages due to something like duplicate content or some other reason. Maybe they lost a lot of their links that got those pages indexed in the first place.
Hmmmm, I hardly doubt that the API shows what will happen when G! updates it's their live servers index. Although I could be wrong..
yfs1 may be right ... on a few domains now ... i get coop weight loss, and weeks later, G starts whacking me with dup content filters. i used to be able to revive them ... it's getting tougher now though.
Then how do you PR? The only way I know how to do this is with a call to www.google.com/search if there is some way that doesn't breach TOS, I'd love to hear it.