Brightscope, a case for developer-friendly open government data

You may recall that I wrote a short article about BrightScope and how its use of open government data, in this case PDF files delivered on CDs on a regular basis, led to a profitable new business, generating jobs in the process. For me, profitable businesses that are generating jobs are a success. However, I wanted to understand the cost to BrightScope to work with government data in this format (PDF files on CD). This cost gives us a benchmark for what other businesses must be prepared to spend to create businesses based upon data in these non-developer-friendly formats.

According to a private DM session I had with BrightScope they have spent approximately $1 million to date, and continue to spend roughly $40,000 per month to work with this data. BrightScope has been in business for a little more than two years, spending roughly a half million dollars a year to process this data. How many new businesses, much less existing businesses, can afford to invest that much money on new products that MAY result in a profitable new venture? Very few.

If, however, this data was already in existence in developer friendly formats like XML, CSV, etc…., the cost would clearly have been much lower. While BrightScope had no guess to share with me, I would estimate a worse case scenario of 3 engineering months (60 days) to build the process up front, a couple of days a month to handle any errors processing this data. If we assume a cost of $2000/day of engineering time the total cost, in two years, would be roughly $168,000 dollars total.

You read that right… The cost to a business would go from roughly $1 million over the course of two years to $168,000 dollars. The ongoing cost would be $4000 per month vs. $40,000 per month. How many more businesses could be created? How many new jobs would there be?

John

Originally posted on Government in The Lab.