Protest challenging agency’s cost, technical, and experience evaluations is sustained where the agency failed to adequately document it evaluation conclusions. Nothing in the record showed that the agency had evaluated the awardee’s proposed direct labor rates. Additionally, the record only showed that the agency had evaluated a small fraction of the activities specified in the solicitation. The agency failed to explain in the evaluation why it accepted the awardee’s deviation form a baseline specified in the solicitation. Moreover, the agency failed to adequately document its evaluation of the awardee’s references under the solicitation’s experience factor.
The Department of Health and Human Services, Centers for Medicare and Medicaid Services (CMS) issued a solicitation for award of a task order for beneficiary oversight and claims review services. Ohio KePRO, Inc. and Livanta, LLC submitted proposals. CMS awarded the task order to Livanta, finding that it had a higher-rated, lower-priced proposal. KePRO protested alleging that CMS botched the cost realism, technical, and experience evaluations.
KePRO first argued that the cost realism evaluation was flawed because CMS only analyzed nine of the 25 labor categories proposed by Livanta. GAO didn’t see this as a problem. The nine categories that CMS examined represented 86% of Livanta’s total proposed labor hours. This was a large enough sample for the agency to assess Livanta’s direct labor costs.
But KePRO’s other cost realism arguments were more compelling. KePRO contended that the cost realism analysis was flawed because it appeared that CMS had only evaluated the reasonableness of Livanta’s direct labor rates. GAO agreed, finding that nothing in the record showed that CMS evaluated Livanta’s direct labor rates. Instead, despite the solicitation’s instructions, the evaluation focused only on reasonableness.
In response to the protest, CMS submitted a declaration from the contracting officer who claimed that he had evaluated the realism of Livanta’s rates and that the agency had found that any rates more than 20% below salary.com rates were considered unrealistic. GAO questioned this contention, noting that 4 out of the 9 categories proposed by Livanta were 20% lower than the salary.com rates. Yet, CMS did not make any adjustment to Livanta’s rates, and the record lacked any explanation as to whether these proposed rates were realistic.
The contracting officer also alleged that he relied on a cost analysis prepared by CMS’s financial management specialist. But that specialist only analyzed whether costs were allowable and reasonable. What’s more, the specialist’s analysis relied on a different sampling than the costs specialist’s analysis. To the extent the contracting officer relied on the financial management specialists analysis of realism, it was unreasonable.
KePRO also contended that the realism analysis was flawed because Livanta proposed significantly lower costs than KePRO, and that CMS failed to account for this difference. Once again, GAO agreed with KePRO. The solicitation required offerors to propose costs for 21 activities. Seventeen of those activities involved non-review work—e.g. transition costs, training, and claims intake. The record reflected no meaningful review of whether Livanta’s proposed costs for non-review were realistic. CMS appeared to have only evaluated Livanta’s proposed costs for one out of the 17 non-review activities. The record was devoid of any substantive discussion of how Livanta’s proposed costs for the non-review work represented the likely level of effort for those activities.
Indeed, GAO noted that given the problems with the non-review activities, Livanta’s proposal should have been downgraded under the technical factor. Under the technical factor, CMS was supposed to evaluated the extent to which offerors’ staffing and level of effort aligned with the SOW requirements. Because the evaluation of Livanta did not assess the vast majority of non-review work, GAO held that KePRO’s challenge to the technical evaluation should also be sustained.
Next, KePRO argued that CMS unreasonably accepted Livanta’s deviation from the solicitation’s baseline level of effort for claim reviews. Livanta had proposed fewer hours than the baseline based on its recent experience performing claim review for CMS. But GAO found that Livanta’s previous review work was different in scope than the work required by the solicitation. The record did not provide enough for GAO to determine whether CMS had considered this difference in scope. Instead, CMS had simply made conclusions about the difference in scope without any analysis and thus failed to adequately document this portion of the evaluation.
Lastly, KePRO challenged the evaluation of Livanta under the experience factor, alleging that the company lacked the experience required by the solicitation. Again, GAO found that the experience evaluation was not sufficiently documented. There was no indication that CMS considered the size of Livanta’s previous projects, and the evaluation did not discuss the scope and complexity of Livanta’s references.
KePRO is represented by Michael D. McGill, Kristen E. Ittig, and Thomas A. Pettit of Arnold & Porter LLP. The intervenor, Livanta, is represented by Alexander J. Brittin, Brittin Law Group, PLLC as well as Mary Pat Buckenmeyer and Jonathan D. Shaffer of Smith Pachter McWhorter PLC. The agency is represented by Lucy G. Mac Gabhann and Douglas W. Kornreich of the Department of Health and Human Services. GAO attorneys Evan C. Williams and Amy B. Pereira participated in the preparation of the decision.
