The typical response from the government contracting incumbency is that the biggest issue holding back technical performance is poor project management – the current vendors are good and can do their jobs, but the program managers inside of government constantly derail their sophisticated approaches with changing expectations, political scheduling, and lack of knowledge. If we only had better people in government, the idea goes, the smart contractor – would be able to do our jobs well.
And there’s a point to be made there. We do need more talent inside of government. It’s hard for government to compete with equity – with pay scales in government capping out at about 130,000, it’s hard for government to hire and retain a good programmer, much less a CIO – unless they’re drawn to serve for non-monetary reasons.
But having consulted for hundreds of institutions, large and small, and governments at every level, the idea that this problem is somehow unique to government is baffling. Bad clients exist everywhere. Part of being a good consultant means being able to work with clients at all levels of knowledge, and even teaching them how to be great clients to work with.
Blaming the client is the oldest trick in the book. It’s toxic. But in the case of Healthcare.gov, it wasn’t legislative requirements that made your code say “<a herf=…” instead of “<a href”, it wasn’t the client that made the consultant leave placeholder copy in the source code of the website, and it probably wasn’t the client who neglected to start the whole project by writing tests that fail.
Yes, some of the problems are cultural. Of the recommendations we make – only two (change pay scales for developers, and make it easy to do technology purchases for less than half a million dollars) involve changes to regulations. But we think the primary problem is competition – that it takes months of process negotiation for a company to get a check, that some governmental bodies make companies do all the work up-front before collecting a check (thereby locking out all but the largest businesses from competing), and avoiding competition altogether by buying things off of pre-negotiated contracts all favor incumbency. And incumbency is the enemy of innovation.
Here’s a good example of what the constituent management system looks like for much of the United States Senate:
It’s built by Lockheed Martin, and costs, from what I can see from Congress’s expenditure reports, it costs thousands of dollars.
Here’s the software used by many of those same Senators – but by their campaigns in order to get elected, not by their offices in order to govern:
That’s built by a company called NationBuilder. It starts at $19/mo.
The reason why your member of Congress isn’t using the same tools they use to run the government as they use to get elected isn’t because of poor project management, or not enough technical know-how in government. It’s because companies like NationBuilder can’t even get in the door.
Of course, that’s but one example – and it’s unusually focused on the legislative branch of government, not the executive (they have different procurement rules) but there are examples throughout government of these kinds of issues and these swaths of disparities between the private sector and the public sector.
The most dangerous part of this problem is that disruptive and talented companies like NationBuilder have been locked out of the federal market for so long that the federal IT contract decision-maker is becoming ever-more disconnected with language, availability and prices available for this level of technology.
That’s why solving this problem with better brainpower and project managers inside of government is shortsighted. Should there be better brainpower and project managers inside of government? Absolutely – we want a government that attracts and retains the best and the brightest, and we can always do better. But getting Tim Cook, Jeff Bezos, Mark Zuckerberg, and Sergey Brin to run Federal IT won’t solve any problems if their hands are tied by the anti-competitive environment that allows for contractors to collect checks for websites like this one.
We need to focus instead on improving access – making it simple for a small business to sign up and contract with the federal government. Making governmental required terms of services for Software as a Service businesses known and available. Making it easy for those SaaS businesses to raise-their-hands and showcase their technology to government buyers, and making it easy for contracting officers to get up to date, private sector pricing information. Those solutions are the ones that will cause drastic change in the way government approaches the way it builds technology.