Version 1.0 of Covered California had to handle 1+ million users, in the extremely complicated world of health insurance, while dealing with extremely sensitive data.
I'm sure it could have been built for a heck of a lot less, but I doubt it's quite as wasteful as you might imagine.
With at least $86 million earmarked for ads, Covered California is counting on global marketing firms to get the job done: Weber Shandwick and Ogilvy Public Relations. According to Covered California, ads will appear in multiple languages on TV, radio, newspapers, billboards and social media.
http://www.bakersfieldnow.com/news/local/California-could-se...
EDIT: It looks like they also opened 3 call centers, with about 400 operators, with plans for a 4th next month with 150 more.
http://www.latimes.com/business/money/la-fi-mo-california-he...
But even allowing for $86 million in advertising (which is still insane considering that the entire Obama campaign spent a little less than a billion -- and that was for an entire country and that money did not all go towards advertising. So $86 million for a single state? That number itself is mind-blowingly ridiculous. Allowing that the $86 million came from the $313 million, we're still looking at over $200 million to build a software application. I'm pretty sure the TurboTax guys didn't spend that much on their first application -- and they're dealing with 50 states plus the federal government's tax laws that change every year.
I just wonder how many uninsured people in California could have been covered with that $313 million they've spent building a web application that's trying to help the uninsured. There are apparently 7 million uninsured in California (the highest number and percentage in the country) so that comes out to about $500 per person. Considering that there will not be a 100% insurance rate in California, it seems like that money spent to build a software application could have covered a pretty high percentage of those uninsured. I'm obviously simplifying, but it seems like the goal of helping the uninsured took an expensive, wrong turn somewhere. That 313 million doesn't include the ongoing salaries and expenses of actually running Covered California. The amount of money spent is staggering considering that not a single dollar of that is actually going to pay any healthcare -- it's only a service to help people find insurance.
[1] http://www.latimes.com/business/money/la-fi-mo-california-he...
http://www.healthexchange.ca.gov/Pages/FinancialReports.aspx
Looks like $120 million for the 4 call centers, $100 million for outreach and advertising, $50 million "other", and $30 million "eligibility and enrollment". Maybe the enrollment system is in that $30 million?
Actually, government organizations are obsessively concerned with costs. So much that, in the interest of controlling costs, they create extensive bureaucratic processes design to assure that every step of an IT system implementation -- and separate rules for every step of any project, IT or otherwise, that involves contracting out -- is necessary, that alternatives have been considered, and that the most cost-effective choice for that step has been taken. And elaborate reporting requirements to all the various control agencies designated to assure that these things are done properly.
Now, there's a couple of perverse results of this: 1) The oversight mechanisms are themselves expensive, and 2) The procedures are mostly old, the IT-specific ones often based on 1970s models of IT with additional layers of accountability but no reexamination of the basic underlying models in the intervening decades, and the general contracting-out ones not well adapted to the specifics of IT.
As a result, the constraints imposed due to cost concerns actually often serve both to create expensive bureaucratic overhead for government IT projects, but also to force the project to be big-up-front-design that gets locked in by contract, with all the well-known costs and risks that come with big-up-front-design of software systems.
Its not that government is not concerned with costs, its that the people in government making decisions about how to address cost concerns don't understand what drives costs in the area of IT (the same may apply in other domains, but that's neither here nor there.)
But you're correct. Government organizations don't care -- I just wish more people could make them care!
Many of the bureaucratic rules that increase costs are the direct results of the voters getting aware of costs and demanding that something be done to constrain them.
If you want the voters to get a deep understanding of what specifically drives costs in IT (or other particular domains of concerns), that's a nice wish, but its not particularly realistic.