18F’s micro-purchasing procurement hack

Using a well known acquisition practice — but one rarely used to buy IT services​ — 18F is going to experiment with new open-source procurement hack​.

Using a well known acquisition practice — but one rarely used to buy IT services — 18F is experimenting with a new procurement hack.

The General Services Administration digital team will begin using the Federal Acquisition Regulation’s so-called “micro-purchase” authority to let software developers bid for work on an open source coding project related to its CALC tool starting Monday. Instead of a monthslong procurement process for the work, 18F should have the work delivered in just a couple weeks for less than $3,500, the threshold for a micro-purchase.

Typically, the micro-purchase system, employing government-issued credit cards, is used for acquisitions of small items like office supplies and equipment or travel tickets. 18F is just expanding that to software development.

Phaedra Chrousos, associate administrator of GSA’s Office of Citizen Services and Innovative Technologies and 18F, told FedScoop that, in theory, the experiment is advantageous compared to traditional procurement for several reasons, “but the most obvious from the government’s standpoint is speed. In this first experiment, we will go from requirements to delivery in less than 15 business days.”


On Monday, 18F will tag an issue on its GitHub repository for CALC, “a tool to help contracting personnel estimate their per-hour labor costs for a contract, based on historical pricing information,” the repository says. Bidding will start at $3,499, and the lowest bidding developer will have 10 days to deliver code satisfying 18F’s given criteria.

But speed is just one reason 18F thinks this procurement hack could benefit its agile delivery of digital services.

“Beyond that, under the authority, we can engage with a broad and diverse vendor pool, including vendors who otherwise have never worked with the government,” Chrousos said. And because the code is open source, “it lets vendors meaningfully evaluate their ability to decide whether to bid or not,” she said. “… [T]here is no guesswork about risk. If we had closed source software, there would be no meaningful way for vendors to evaluate how difficult it would be to get the job done on budget.”

And on the private side, Chrousos said, this model lets small and nontraditional vendors compete for government contracts — and “contribute to the American public and get paid for it.”

For this first iteration, only businesses registered on GSA’ System for Awards Management can participate. But, “if the experiment works, we may look to expand beyond registered companies,” 18F’s David Zvenyach wrote in a blog post.


And there is always the chance it doesn’t work; 18F recognizes that. Zvenyach even wrote on the post: “[T]his might be a terrible idea.” But that’s what insiders say is special about 18F — if they’re going to fail, they’d rather fail fast, with little risk — and learn from it.

“We’ll try it and see what happens,” Zvenyach wrote. “If it works, that would be fantastic. If it doesn’t, it’ll be an inexpensive experiment and we will have learned some new things.”

Latest Podcasts