The rise of the microconsultants

One of the first -- and even more than two decades later, still one of the most important -- changes during the procurement reform era of the 1990s was the introduction of a “micropurchase” authority in the Federal Acquisition Streamlining Act of 1995. In the background was the desire to allow program customers the ability to use a government credit card to make very small purchases directly, without the significant extra time (think of the purchase order request landing at the bottom of the contracting person’s in-box) and administrative cost that come with working through the contracting office.

An interesting backstory to this statutory change was the role of a young contract specialist at the Department of Health and Human Services named Michelle Craddock in helping develop the administration’s legislative proposal on this. Even before the 1995 law, the government could make purchases under $2,500 without going through a competition. Having just arrived in Washington in 1993 to become administrator of the Office of Federal Procurement Policy, I was tasked with trying to find out if there were any statutory barriers to using a credit card for under-$2,500 purchases. On a learning tour of contracting offices to prepare me for my job as a Washington outsider awaiting Senate confirmation, Craddock was one of the contracting people I met in a visit set up by the HHS senior procurement executive.

When I asked Craddock about things I might do to improve her ability to do her job, she mentioned there was a big-box computer store across the street, but HHS couldn’t use it because there were a number of contract clauses implementing various procurement policies that had to be included for all procurements, even those of the smallest dollar value. (Buy America restrictions were the most significant of the bunch.) Could we eliminate those clauses for contracts under $2500, the competition threshold, she asked?

Neither I nor the staff had thought about this problem before, and when I returned to my office I requested that team look to see what such clauses existed. It turned out there were seven. A $2,500 threshold for each we put in the legislation , and this made it into the law. I told Craddock at the time that if the bill became law, we would invite her to the signing ceremony, which we did, and both President Clinton and Vice President Gore called out Craddock in the Rose Garden ceremony. (Kudos to Mat Blum from the Office of Federal Procurement Policy for finding the C-SPAN video of that ceremony.)

Under the moniker “micropurchase threshold,” the authority granted in 1995 brought about a huge expansion in use of the government credit card, a big quality of life benefit for frontline federal employees involved in buying things for their offices. It was used for all sorts of purchases, frequently personal computers and peripherals. But for its first decades, it has been used almost exclusively to buy products, not services, which appeared to require larger contracts than $2,500.

Until last December, there had been no governmentwide statutory changes in the original 1995 micropurchase threshold, though the threshold had risen in two stages to $3,000 and then $3,500, based on a regulatory ruling that tied most thresholds for various procurement policy requirements to inflation. In 2016, the defense authorization bill subsequently increased the threshold for DOD to $5,000 and then last year’s defense authorization raised the threshold for civilian agencies to $10,000 (while, bizarrely, keeping the one for DOD unchanged at $5,000).

We are still in a transition period -- implementing regulations are expected by the end of the summer, before which agencies are not allowed actually to use the new thresholds, but the Department of Veterans Affairs and Department of Homeland Security have received a class regulatory deviation to allow them to use it now, and the General Services Administration has published guidance to other agencies on how to request deviations for themselves.

What should have been the headline news from the increase in the micropurchase threshold last December got virtually no attention at the time. Raising the micropurchase threshold to $10,000 makes it practical for the first time to use micropurchases to buy services, and not just products. This has the potential to be revolutionary for the procurement system.

During this transition period before the implementing regulations are out, some fast movers in government and industry have put on their thinking caps and gotten busy asking what useful services can be bought for $10,000. First out of the gate has been the VA, under the leadership of Mark Junda, a creative contracting professional whose work I have discussed previously.

VA become the first government agency to use the $10,000 micropurchase authority by awarding two orders to make recommendations for various ways the VA could more effectively use APIs to help the developer community, and to recommend priorities for new APIs for the department to develop.

The regulations for micro-purchases do not require getting any competition at all, but VA officials decided to make the process competitive and transparent, in a streamlined way. They have tried to be careful in a solicitation to make clear what the government wants, and what the acceptance criteria are. Proposals are limited to one page, and evaluation is therefore rapid and pretty painless, with no requirements for how proposals must be evaluated –- a feature that appeals both to government and vendor.

Just a year ago this week, Chris Cairns, a longtime veteran of 18F who had then recently left the government, announced the formation of a new firm called Skylight one of whose specialties would be what he called “microconsulting.” 

"Not every problem requires a full-blown consulting engagement,” he wrote in the announcement on LinkedIn. “Often times, a short burst of work on a specific topic is all that's needed to help propel government managers and teams forward."

Cairns developed the idea that became microconsulting while running a consulting unit at 18F. He observed that agency demands for 18F services were far larger than the organization’s capacity. In trying to figure out what jobs to take, he was influenced by learning that many government customers believed a short engagement of only a few days could provide most of the value agencies sought.

Skylight is the contractor on VA’s already-competed microconsulting contracts. Cairns' company now offers a microconsulting “health check” for agency digital projects: to “identify high-impact opportunities to improve the health of your digital service delivery project." According to Skylight's description, "We rapidly assess key dimensions, such as the user experience and code quality, and recommend priority areas for improvement.” Other possible uses for microconsulting Cairns mentioned to me are developing hiring plans for certain agency jobs and making recommendations for Section 508 compliance.

There is now a second small firm in this space, Arc Aspcio, founded by Lynn Ann Casey, formerly of Accenture. (Check their website to see the origin of their unusual name!)

Arc Aspcio recently organized an ‘IdeaSlam” to discuss ideas for microconsulting projects. Ideas presented included taking some data, visualizing it in PowerPoint, Tableau, or some other tool to show the key findings from the data and then we run a decision analysis meeting using the visualization to help executives see the findings so they can make a decision. Another idea was to provide services to support social media use during a crisis, where the deliverable could be a social media communications plan, a summary of channels, with specific recommendations on the best channels and the messages for communication. 

A number of projects discussed involved specific elements of developing an RFP or other kinds of acquisition support. Cairns believes that even some simple software development projects can be done for under $10,000. 

Beyond one-off engagements of these kinds, another important potential use of microconsulting is to procure larger efforts iteratively with a series of microconsulting contracts, where each contract represents some incremental progress, in the spirit of agile software development.

There are other attractive features of this process for both the government and the vendor. For example, the micropurchase regulations allow for government evaluation of contractor past performance with no formal systems, data gathering or appeals, making it easier for a new contractor to develop a past performance record. Cairns suggests that new vendors may price their offerings particularly aggressively to increase their chances of being selected and giving them the opportunity to develop past performance (a practice that frequently occurs in the commercial world when contactors are trying to enter a market against incumbents).

Shortly after Cairns’ original 2017 declaration, I blogged about this, saying it might disrupt the existing world of large contracts with large teams working on large projects. At that time, I thought of the potential disruption in terms of cost savings for government: “Traditional large contractors, “facing a less-competitive landscape,” I wrote then, “may be likely to provide drawn-out consulting advice from a bevy of staff rather than find ways to provide such services with much lower level of effort.”

The culture of the traditional system assumes that any one project will be lost in a sea of other work, creating a culture that doesn’t value saving money. By contrast, projects that are born small create a cost-savings environment. Microconsulting is likely to disrupt by putting pressures on traditional contractors to deliver with fewer bodies and less time.

I still believe this is a big plus with microconsulting, but now another benefit looms just as large: encouraging new entrants into the federal marketplace. This is an amazing opportunity for new vendors. They can get small jobs with very small proposals, with no requirements for various certifications or other special requirements for contracting with the government, and, under the micropurchase regulations, a vendor doesn’t even need to be registered in the government’s System for Awards Management (SAM). It is also possible to do a succession of small jobs in a similar domain, making the revenues one can achieve for a small new firm not- insignificant.

The flip side of this is that micropurchases might be the only way for a new contractor to enter the federal marketplace, at least as a prime. Few agencies would be willing to take a chance on giving a larger contract to a vendor with no past performance; micropurchases represent a digestible bite for agencies.

Cairns also notes other advantages to microconsulting -- that it provides an opportunity for narrow niche specialists to get work that might be harder to get as part of large traditional contracts. And in the gig economy, some professionals find the opportunity to work on a larger number of smaller projects more attractive than being tethered for a long time to big ones. Finally, it is typically easier to explain a small requirement clearly than a larger one.

“For a $25 million job, it’s hard to find definable criteria," Cairns said. "I n a microconsulting solicitation, it is easier to make it clear to me as a vendor what they are looking for.” This reduces miscommunication and improves performance.

One big worry that agencies are likely to have about using microconsulting is that agency administrative costs for doing many small procurements will go up compared with doing one much bigger one. The most-legitimate worry is about the cost of developing 10 separate requirements for 10 microconsulting efforts in a similar area.

However, if the alternative is developing a single very detailed requirement, as in waterfall software development, versus the less-detailed, iterative requirements of agile, the cost of 10 iterative requirements, each of which has only incremental changes from previous ones, may be less than the cost of developing one big procurement. Alternatively, under an existing big contract, the government may in fact develop no real requirements for the big chunk of work, just get the contractor started with vague direction, and provide ad hoc changes in direction as the work proceeds.

The administrative costs of proceeding in this way might be less than a more formalized development of 10 specific requirements, but one should not exaggerate the difference. In either case the government needs to spend resources considering what different things it will be asking the contractor to do over time, then negotiating these with the contractor. And there are of course benefits in terms of the likely quality of the work to being more specific.

As for the administrative costs of proposal evaluation, if the agency follows the micropurchase language of the FAR, it does not even need to receive or evaluate proposals. If an agency chooses to get proposals (which I suspect will increasingly occur as microconsulting spreads), evaluating responses to 10 proposals of the sort the VA has been getting will likely cost less than evaluating one large proposal under the FAR, including preparing reports for a source-selection official.

This is just getting started, but I predict we’ll be hearing a lot more about microconsulting. This could be a big deal.

Note: This article was updated on July 12 to add that DHS also has received a class regulatory deviation.