Site visits random? No.

The official line has always been that the BearingPoint Extended Outreach Site Visits are randomly assigned based on recently paid invoices. There is supposedly an algorithm that is randomish, but takes into account that some FRNs are invoiced more frequently than others, etc. If such an algorithm exists, it isn't working.

Recently, I was browsing the latest quarterly report from BearingPoint to USAC, and the bottom of the table on page 8 jumped out at me. It breaks down how many site visits are done for each Category of Service. What struck me: more than half of all Site Visits are for Internal Connections FRNs. So I went back to my database of all FRNs and looked at what percentage of funded FRNs are Internal Connections. In 2004, it was 13% of funded FRNs. In 2005, 10%. If you look at the dollars disbursed, it's closer in 2004: 41%. In 2005, though, it dropped to 23%. Meanwhile, telecom FRNs are under-visited: about a quarter of all visits, but a two thirds of all FRNs (and half of all dollars). You can see the tables here.

I think this is good. Internal Connections is the biggest source of waste, fraud and abuse is, so that's where SLD scrutiny should be focused. Site Visits are a good tool for reviewing equipment purchases, because they can see that the equipment is actually installed and working.

But the SLD should come clean: the Extended Outreach Site Visits are being used as audits, and they are targeted at Internal Connections FRNs.