Scraping, by its very nature is far more bandwidth and resourse greedy than quering the API. Those currently using the API have in the past scraped, and will more than likely return to doing so in the absense of a functional SOAP API. Google knows this and thus this renders both the bandwith and commercial argument for this move as less stable.

From a practical perspective, the API has is recent months returned increasingly different results to the SERPs… so much so that they are useless. Due to the new datacentre architecture, it is reasonable to speculate that the API is now redundant…

1. Paring down of services – they got rid of Google Answers for the same reason. I am sure that the thousands of SEO tools out there which used the Google API probably cost them a bit in terms of bandwidth and hardware.

2. Index Security – We know that Google resists the idea of people manipulating their SERPS. In their view, they are providing the best results without pesky SEOs coming in and throwing them off. Well, the SOAP API allowed great freedom programmatically, allowing thousands of SEO tools to use the listings for competitive intelligence. I am sure that Google considers it to be in their best interest to make the life of an SEO difficult at best, and disabling the SOAP API has virtually neutered a large number of the tools that newer SEOs depends on for research.