Some devoted Census Bureau data users have recently posted concerns about American FactFinder and its cost. We’ve had a lot of this feedback since it launched late last year and many improvements based on that customer feedback have led to a steady increase in satisfaction with the site.

It is important to note that FactFinder is not just the website you see, it is also a massive tabulating operation.  The back end takes individual responses to the 2010 Census, removes personally identifiable information, and adds them together to produce the statistics that summarize how our nation is changing.

FactFinder is part of a continuing effort to take information developed on specific forms for specific purposes like redistricting or export administration, combine it and make the information we produce as broadly and easily available as possible.

The cost of all this includes specially purposed hardware, with 200 servers capable of handling 2 trillion cells of data containing over 375 billion individual data cells adding up to 11.5 terabytes of data.  On top of that are the development costs for applications and testing, as well as IT security and user training. In addition to data from the Decennial Census, Factfinder contains data from the American Community Survey, the Economic Census, a number of annual economic surveys, and the Population Estimates Program (a total of almost 30 datasets).

We continue working to make FactFinder more intuitive and user-friendly, with another set of improvements scheduled for January .  The American Customer Satisfaction Index scores for FactFinder mirror the comments from our friends in the blogosphere, with our data getting higher rankings than functionality and navigation.  But as we make improvements, those functionality scores are rising. .  We expect FactFinder will continue to have heavy use, currently 600,000 to 700,000 unique visits a month.

Going forward, the Census Bureau is very focused on developing tools that serve the many needs of our data users, from a student looking for a simple fact like her hometown population, to a power user who wants to download a massive national data set, and everyone in between.  Working in collaboration with other statistical agencies, academia and the private sector, we want to lead the way in making information more transparent, accessible and easier to use.

The imminent release of our Application Programming Interface, commonly known as an API, will let developers build apps tailored to meet the needs of a variety of audiences.  (Thanks to Rebecca Rosen at The Atlantic and Josh Barro at Bloomberg for the API shout-outs.) We’re looking forward to seeing what you come up with.  The result will help fulfill our mission of getting the highest-quality data to empower Americans to meet the needs of our changing, growing nation, in ways we can’t foresee.