LCAT Math
Last week, we talked about how a challenger (KPMG LLP) managed to beat an incumbent (Guidehouse) in a procurement with the Office of the Director of National Intelligence's Chief Financial Officer. We observed that KPMG won because its proposal scored higher than Guidehouse across every factor, despite having a significantly higher price bid. As I mentioned at the end of that post, however, winning the award was not the end of the story, because Guidehouse won the protest at the Government Accountability Office. Today, we cover why that happened.
But first, we need to talk about "labor category mapping". Labor categories—known in the biz as "LCATs"—are a way of labeling the work that people are doing under a services contract. LCATs can be everything from "Cloud DocuSign Developer" to "Executive Strategic Advisor" to "Cement Mason".[1] Stated another way, a LCAT is the government-y method of describing—in the words of Richard Scarry—"What Do People Do All Day?"
When a company prices its services, the company will typically set different compensation levels based on the education and experience of the individuals in that labor category. For example, a company might have an LCAT for "Accountant I" that requires a bachelor's degree and no work experience and an LCAT for "Accountant III" that requires a master's degree and 5 years of experience.
The trouble, though, is that every company is slightly different. Different firms have different labor categories with different prices. Government procurement rules, as you can image, demand an even playing field across competitors.
So, what ends up happening is that the government will establish its own LCAT expectations and require vendors to "map" their LCATs to the government's. We've talked about this practice—labor category mapping—before:
Contractors map their employees to the government's defined labor categories. And the variables that contractors use to map the labor categories to the government's definition are the nature of the work (can a mechanic do carpentry?), education (do they have a degree?), and the years of experience. This mapping is an exercise that exists to make sure that the government actually gets the skills they need, without overpaying.
In the case of Guidehouse, as part of the solicitation, the government defined 9 labor categories, with different qualifications for each level, within each labor category. Each LCAT had educational and experience requirements. Relevant to the protest, 8 of 9 LCATs were all "expert" roles, including "business intelligence expert", "financial management expert", and "financial counterintelligence" expert.
You might guess where this goes:
According to Guidehouse, “KPMG took exception to the requirement to provide each labor category in SOW Appendix A by failing to propose three of the required labor categories”: business intelligence expert, financial management expert, and communications expert. Guidehouse asserts that, “Instead, KPMG proposed only ‘mid-level’ individuals” for the categories.
ODNI responds that, essentially, Guidehouse is elevating form over substance. The agency does not dispute that, for the business intelligence, financial management, and communications categories, the awardee proposed personnel with a skill level labeled “mid-level” or “junior.” According to the agency, however, KPMG’s skills matrix evidenced that, labels aside, the awardee offered personnel “who far exceed the requirements for an ‘expert’ under SOW Appendix A” and offerors “did not have to propose individuals labeled as experts; they ‑‑ like every other offeror – had to propose individuals who met the Agency’s requirements for what constituted an expert.”
Specifically, in response to the protest, ODNI asserts that the solicitation allowed offerors to propose individuals that met the SOW’s requirements or offered “equivalent levels of educational backgrounds and experience.” The contracting officer then offers a chart that purportedly “demonstrates how each of the proposed personnel met the requirements for the three” labor categories at issue.
Essentially, Guidehouse argued that KPMG didn't offer experts; they only offered junior or mid-level individuals, and therefore didn't meet the agency's requirements. Meanwhile, the agency argued that, even if KPMG didn't technically label their individuals as experts, the proposed personnel were qualified enough.
Which would be a good argument, I guess, if the agency actually evaluated the expertise of the proposed personnel. As GAO explained, however, that didn't happen:
The problem, as Guidehouse points out, is that the agency has produced no evidence that it contemporaneously evaluated KPMG’s proposed labor to determine whether they met the agency’s requirements under SOW Appendix A or equivalent qualifications.
Making things worse, when the GAO independently reviewed some of the proposed personnel's credentials, they were a bit wanting:
For example, SOW Appendix A provides that a business intelligence expert must, among other requirements, have earned “a Bachelor’s degree in Business, The agency contends that KPMG’s two proposed personnel met the requirements, but then cites candidate A, with a master’s degree in criminology/applied data science, with no reference to or explanation of any bachelor’s degree (or a master’s degree) in one of the three specified fields. A business intelligent expert also requires a minimum of 5 years of experience in financial management disciplines. The agency again contends that KPMG’s proposed personnel met the requirements, but does not address that KPMG’s proposal claimed only “[o]ver 3 years of experience in financial management” for candidate B.
KPMG lost here because its "experts" appear to have been too junior and the agency didn't check their credentials. Womp womp.
Still, three lessons about LCATS emerge from this fact pattern: one lesson for the government, one lesson for industry broadly, and one for Guidehouse.
The government's lesson is that, if you want a vendor to have flexibility in defining LCATs, you either need a wide spread or you need to document any exceptions. If you're going to put a minimum experience requirement or a minimum education requirement in the solicitation, you're also going to need to actually check the proposal's resumes and do some basic addition.[2]
The industry lesson is that, if you're going to bid out certain individuals for work, you need to actually check their resumes to make sure they meet the requirements. This advice feels a bit like making sure your SAM registration doesn't lapse. But folks, I really want to stress how important this is. If the government says "5 years experience of financial management disciplines" and your employee's resume has 4 years of experience as a CFO and 1 year as a frozen banana stand operator, you better talk about how your employee checked receipts or whatever.[3] When you are doing LCAT mapping, make sure the math checks out.
And the lesson for Guidehouse? Even though you won the protest, consider a postmortem on your capture effort here? Your argument is that KPMG shouldn't have won because they were too junior, which is fine and good. But recall that KPMG's technical approach was "exceptional" where Guidehouse's approach was only "very good". Which suggests that even potentially unqualified personnel were more seen as more favorable than the incumbent's team. Ouch.
And that brings our two-part story to a close. If you have a good proposal strategy, you can beat an incumbent. But even if you nail the proposal's win themes, you need to make sure you're checking your boxes on your compliance matrix.
That's how this game works, folks. Them's the breaks.
[1] I spent minimal effort looking through price lists to find accessible examples of real labor categories that spoke to me. But some LCATs be wilding. Supposedly there's an LCAT for "captain of industry" out there on some IDIQ? I strongly encourage folks to email to tell me about their favorite LCAT.
[2] The GAO also sustained Guidehouse's argument that ODNI incorrectly evaluated their proposed program manager. The government dinged Guidehouse because it believe that the PM didn't have the requisite years of experience, largely based on the job titles that the PM previously held. But GAO concluded that "the evaluators’ stated basis for the weakness--a lack of information supporting X’s claim to program manager experience--does not reasonably account for information provided within Guidehouse’s proposal describing X’s duties managing programs for federal agencies." So there's that, too. In addition to math, the agency's have to do more than just read the job title.
[3] Or emphasize, as Arrested Development fans out there know full well, there's always money in the banana stand.