GSA's blockchain blockbuster

The other week I got an email from Jose Arrieta, the new director, arrived less than a year ago, of the General Services Administration's Schedule 70 (IT products and services) operations. He told me his organization had awarded in late June the federal government's first contract to use blockchain technology, involving moving FASt Lane, a service that Schedule 70 designed to speed how fast new vendors can get on schedule, onto the blockchain. Might I want to write about it, he asked?

Blockchain was initially developed as a platform to allow trading of the cryptocurrency bitcoin in a trustworthy, transparent environment. Bitcoin has a bit of a checkered past, and even something of a checkered present, as a play-space for libertarian activists who didn't like the government's control of money, and as a possible vehicle for criminal or terrorist financing. (A Google search of "bitcoin" and "dark web" generates 2.7 million hits.) I suspect that most people in government who have heard of blockchain still associate it mainly with bitcoin.

Recently, though, blockchain has been getting a lot of mainstream attention for possible applications well outside of bitcoin, including several articles over the last few years in the Economist. Both FCW and GCN have reported on the technology with some frequency, and the IBM Center for the Business of Government has posted two blogs on blockchain in government.  

At its simplest, blockchain is a ledger -- a record of transactions and other relevant data about companies or individuals -- that is distributed; its content is shared in real time with anyone with access to the blockchain. Another design feature of blockchain is that entries to the ledger are immutable -- somebody can update an entry on the ledger, but the update only adds to the original entry, it can't replace it (so people can't hide their tracks and erase the past).

It has been quite an education for me to talk with Arrieta and his vendors, Sapient and United Solutions, about what they've been up to, and to read some of the new literature on blockchain.

I started my search by talking with Arrieta about the path that led him to pursue a blockchain implementation at GSA. As I did so, it became clear that he was driven by serendipity and opportunistic search (in the positive sense of the word) rather than some methodical strategic planning process that outlined goals upfront and thought about the optimal way to reach them.

Before joining GSA, Arrieta spent three years with the Treasury Department as its small business executive. He admired his boss, Deputy Secretary Sarah Bloom-Raskin, for her willingness to involve him in senior staff meetings beyond his specific bailiwick. Treasury has a seat on the National Security Council, and Bloom-Raskin asked Arrieta to represent Treasury on a working group on emerging technologies because he had earlier worked as program manager for a Treasury project to use big data in anti-terrorism efforts.

When he was starting work in this group, Bloom-Raskin mentioned to him that a new technology called blockchain might revolutionize financial services. "Her one comment sparked my interest," Arrieta said, and he got blockchain included on a list of technologies the NSC group would examine.

At this point, Arrieta started avidly working to learn more. "As a leader," he says, Bloom-Raskin "was very open to me exploring." In a policy shop, he added, "you have access to lots of experts – I met about blockchain with people at MIT and went to some conferences."

He also discovered that "there were folks in government who were very passionate about blockchain as a mechanism to help the unbanked," both domestically and in developing nations. That use case resonated with Arrieta, whose wife is from Bolivia, one of the poorest countries in Latin America. But "I had a background in contracting," he said, "so I [also] became interested in using this for more than just financial transactions."

Arrieta was clearly motivated by self-directed energy. Trying new things may require passion, and passion is more likely to result from self-generated energy.

The next instance of serendipity came in late 2016, when GSA approached him about applying to be head of the Schedule 70 program. During his interview, he repeatedly raised the idea of digitizing Schedule 70. In early 2017 he arrived at GSA.

Arrieta then approached his boss, Mary Davie (recently named deputy commissioner of the Federal Acquisition Service), to see if he could get $250,000 to do a proof of concept to digitize one part of Schedule 70. He originally suggested laptops and desktops, a growing area for the schedules because of the category management initiative.

Davie told him that it would take two years to jump through internal approval hoops, and that he should take a business process that was under his control. FASt Lane fit that criteria.

The acquisition strategy, developed by Arrieta together with staffers Heidi Mcfall, Michelle White and Nick Adams, structured the RFQ to reduce risk, making early oral presentations a possible kill point for the project if nothing interesting came in.

Davie asked to be kept updated, but Arrieta said she raised neither the issue of bitcoin's checkered past nor the risk of being the first to try something new in government. (Interestingly, when the solicitation hit the street the innovation unit at Treasury offered GSA $1 million to participate, so they could be involved in scaling the pilot -- another example of a larger innovation ecosystem in the government).

The RFQ presented a statement of objectives that called on the contractor to develop a proof of concept for digitizing FASt Lane that would include developing a distributed ledger in two weeks and a capability for automated processing of two of the steps in the FASt Lane workflow in three weeks, using agile software methods to produce a minimal viable product.

"We told offerors that if they couldn't develop the distributed ledger in two weeks, they shouldn't bid." The vendors began working on the proof of concept in July and were done by the end of September. 

In most primers on blockchain, three features are stressed again and again: verifiability, immutability and transparency. At least for blockchain entries that involve a transaction between two parties (such as buying or selling a house), the existence of the transaction on the blockchain itself verifies the transaction. This obviates the need for expensive and time-consuming involvement of intermediaries (e.g., banks or title companies) confirming that your assets are what you claim they are. This creates a powerful new way to create trust.

Immutability also creates trust, because it prevents parties from eliminating or altering information on a ledger to benefit themselves (such as by removing negative information about legal actions).

And transparency is a big benefit of the blockchain for a business process such as FASt Lane that involves the government's interaction with vendors -- all interactions, recommendations, and decisions are stored and viewable. For example, if an offer is rejected, the information affecting that decision is available as evidence for protests and audits; transparency also offers firms access to the information in the ledger about them, so they can see if there are any errors.

When blockchains are used to transform a business process like FASt Lane, verifiability and immutability may not be the central benefits. What they make possible, however, are major speed and cost savings from using blockchain for performing such processes.

Blockchain allows far cheaper, more flexible, and more accurate data manipulation than traditional databases. In traditional systems, one creates multiple databases with specific data elements and operations that can be performed on them. If one wishes to introduce new data elements or new ways to analyze some data, one must modify the database. Multiple databases also create risks for errors in data transfer from one database to another. With blockchain, there is only one shared data layer that includes all data in the ledger, and an open application programming interface connecting any user to the ledger. The applications that leverage the API for any given user's needs can be modified as appropriate without needing to modify a database. Also, that there is only one ledger eliminates the errors that can creep in when transferring data from one database to another.

A crucial element of GSA's proof of concept was to change two steps of the process --the review of vendor-submitted financial statements, used to determine the vendor's financial solidity, and the preparation of a negotiation letter that outlines issues over which GSA wished to negotiate before accepting the vendor on schedule.

On the financial review, the traditional process involved a staffer extracting financial information from the material the vendor provided and calculating the firm's financial health, which took up to a month. The capability developed for the proof of concept accelerated that process so that, in most cases, it's nearly instantaneous. An automated financial review now occurs immediately. Offers that pass are moved onto the next step in the workflow; those flagged for further review or rejected are routed to a human reviewer for further analysis. 

A pre-negotiation letter, meanwhile, is a document prepared for GSA negotiations with a vendor, listing issues to raise in negotiations. The blockchain, with its full history of entries in the ledger, became the system of record for an entire offer, replacing multiple emails back and forth between the government and vendor, and back and forth with the contracting officer checking multiple systems. It has reduced the time to prepare the letter from 15-30 days to under 10 days. (The system does not currently suggest items for negotiation to the government, but in the future will analyze past pricing data to flag proposed prices that are out of line.)

"With the proof of concept, we were trying to create positive momentum for major change through smaller changes," Arrieta told me. "The two areas we chose had created real problems for the contracting officer in the past; these were problems the workforce asked us to tackle."

Through mid-December Arrieta is working to share what they've learned with other parts of GSA and with everyone on his Schedule 70 staff, and then to industry associations. They will also practice with staff running the processes they've redesigned with real data, in preparation for putting the system live.

I also spoke with David Nguyen, CEO of United Solutions, a 25-employee small business that has been prime on the blockchain work, and Alexander Zelenovic, the vice president for public sector strategy at Sapient, a large business IT contractor mainly focused on the commercial sector but with some government presence as well. The two are both Sloan School of Management grads and have worked together in the past.

My question, especially to Nguyen, whose company does mainly government work, was why they would invest in developing blockchain capability in advance of any contract. Lots of work is set aside for small businesses, he explained, and "there aren't many small businesses that know about emerging technologies," so his company would have a leg up bidding for such work. Nguyen felt that justified making an investment in blockchain.

Apparently, his was not the only small business willing to do so. Eleven firms bid on GSA's order -- a nice statement about the appetite for innovation and risk these days among at least some small business contractors.

Arrieta's opening salvo for GSA's involvement in blockchain may be a blockbuster. It is of course impossible to predict how important blockchain will end up being for government, industry, and our society; who would have guessed at the dawn of the internet how much it would enrich and disrupt the world? Arrieta believes blockchain will set off a new industrial revolution, but who can know what uses nobody conceives of today will end up being important?

It is interesting that a number of companies and governments – including, amazingly, the government of the United Arab Emirates – have launched challenges asking people to submit and enter ideas about new blockchain applications. Stay tuned, and be prepared to hold on to your seats. This could be the start of something big.