Skip to main content

Computer World Services Corporation

B-418287,B-418287.2 Feb 26, 2020
Jump To:
Skip to Highlights

Highlights

Computer World Services Corporation (CWS), of Washington, D.C., protests the issuance of a task order to Intellect Solutions, LLC, of Alexandria, Virginia, under request for quotations (RFQ) No. 70Z07919QPNZ00200, issued by the Department of Homeland Security, United States Coast Guard, for centralized service desk information technology support services. CWS argues that the agency misevaluated quotations and made an unreasonable source selection decision.

We sustain the protest.
View Decision

DOCUMENT FOR PUBLIC RELEASE
The decision issued on the date below was subject to a GAO Protective Order. This redacted version has been approved for public release.

Decision

Matter of:  Computer World Services Corporation

File:  B-418287; B-418287.2

Date:  February 26, 2020

Jonathan J. Frankel, Esq., and Karla J. Letsche, Esq., Frankel PLLC, for the protester.
Sharon L. Larkin, Esq., and James M. Larkin, Esq., The Larkin Law Group LLP, for Intellect Solutions, LLC, the intervenor.
Shandra J. Kotzun, Esq., Department of Homeland Security, for the agency.
Scott H. Riback, Esq., and Tania Calhoun, Esq., Office of the General Counsel, GAO, participated in the preparation of the decision.

DIGEST

Protest challenging agency’s evaluation of quotations and source selection decision is sustained where record shows that agency’s evaluation and source selection were not reasonable or consistent with the terms of the solicitation and applicable statutes and regulations.

DECISION

Computer World Services Corporation (CWS), of Washington, D.C., protests the issuance of a task order to Intellect Solutions, LLC, of Alexandria, Virginia, under request for quotations (RFQ) No. 70Z07919QPNZ00200, issued by the Department of Homeland Security, United States Coast Guard, for centralized service desk information technology support services.  CWS argues that the agency misevaluated quotations and made an unreasonable source selection decision.

We sustain the protest.

BACKGROUND

The RFQ contemplates issuance, on a best-value tradeoff basis, of a fixed-price task order under the successful concern’s Federal Supply Schedule (FSS) contract to provide these services for a base period of 5 months and four 1-year options.  The competition was confined to eligible historically underutilized business zone (HUBZone) small business concerns.  Agency Report (AR), exh. 1, RFQ at 3.  The agency was to consider price and two non-price evaluation factors, technical capability and corporate experience.[1]  Id. at 10-12.  As the RFQ did not specify the relative weighting among the evaluation factors, we conclude that they were weighted equally.  See Bio-Rad Laboratories, Inc., B-297553, Feb. 15, 2006, 2007 CPD ¶ 58.  The agency was to evaluate the firms’ total prices for reasonableness, and would evaluate selected unburdened labor rates for realism.  RFQ at 12.  The RFQ provided that the agency would make a best-value tradeoff decision based on consideration of all the evaluation factors.

The agency received four quotations in response to the RFQ.  AR, exh. 25, Source Selection Decision Document (SSDD) at 2.  The agency engaged in discussions with CWS and Intellect and solicited, obtained and evaluated revised quotations.  The agency assigned adjectival ratings of high confidence to both quotations under both non-price factors.  AR, exh. 25, SSDD, at 3.  CWS’s total evaluated price was $56,842,229, while Intellect’s was $47,001,784.  Id.  CWS’s quotation was eliminated from further consideration because the agency found its total evaluated price unreasonably high when compared to the government estimate of $48,642,364, and because one of its unburdened hourly rates was deemed unrealistic.  Id. at 4.  The agency concluded that only the quotation of Intellect remained eligible for issuance of the task order and selected the firm for award; this protest followed.

DISCUSSION

CWS argues that the agency misevaluated both its and Intellect’s quotations.  The firm argues that the agency erred in finding its quotation both unreasonably high in price and unrealistic in terms of one of its unburdened hourly rates.  As for the Intellect quotation, CWS primarily maintains that the agency failed to consider that it offered a level of effort significantly below the level of effort the agency determined was necessary to perform the requirement and offered to perform with a less qualified labor mix.  CWS also challenges the evaluation of Intellect’s quotation under the corporate experience factor, maintaining that one of its references did not meet the RFQ’s experience requirements.

We have reviewed all of CWS’s contentions and sustain its protest for the reasons discussed below.  We note at the outset that, in reviewing protests challenging an agency’s evaluation, we do not independently evaluate proposals or quotations.  Rather, we review the record to determine whether the agency’s evaluation was reasonable and consistent with the terms of the solicitation and applicable statutes and regulations.  McCann-Erickson USA, Inc., B‑414787, Sept. 18, 2017, 2017 CPD ¶ 300 at 3.  While we will not substitute our judgment for that of the agency, we will sustain a protest where the agency’s conclusions are inconsistent with the solicitation’s evaluation criteria, inadequately documented, or not reasonably based.Id. 

Determination That CWS’s Price Was Unreasonable

CWS argues that the agency acted irrationally in finding its price unreasonably high.  The protester, who is the current incumbent for the requirement, argues that the solicitation increases the level of effort called for in comparison to the services provided under its predecessor contract, and increases the credentials required for employees to work on the task order.  CWS argues that its quoted price reflects these realities, and that there was no basis to find its price unreasonable under the circumstances. 

CWS argues that the agency used a flawed independent government estimate (IGE) as a benchmark to measure whether proposed prices were unreasonably high.[2]  The protester maintains that the IGE does not accurately reflect the solicited requirements and was prepared using data that essentially is inapposite for purposes of determining whether the proposed prices here were reasonable.  CWS also argues that the agency’s determination that its price was unreasonably high is further undermined by the agency’s use of its flawed IGE as its only point of comparison in making its reasonableness determinations.  According to CWS, if the agency had compared the quoted prices to one another, it would have found that its price was reasonable. 

Our review of the record largely confirms CWS’s arguments relating to the accuracy of the IGE, and also shows that, because the agency used only the IGE in making its price reasonableness determinations, this may have led to misleading results.  We discuss both matters below.

As noted, this acquisition was conducted using the FSS, and the competition was confined to FSS contract holders that were HUBZone small business concerns.  However, the record shows that the agency prepared the IGE using labor rates from the General Services Administration’s Alliant II governmentwide contract vehicle.[3]  The record shows that the agency appears to have used some sort of averaged hourly rates from this contract vehicle.[4]  AR, exh. 27, Price Analysis Report, at 1.

Given that the competition was confined to HUBZone small business concerns with FSS contracts that include published hourly rates for all labor categories contemplated under the RFQ, AR, exhs., 9, 13, CWS’s and IS’s FSS Contracts, there was no rational basis for the agency’s use of labor rates offered by large businesses on the Alliant II contract vehicle.  The agency states that it originally prepared the IGE using the Alliant II contract vehicle because it intended to meet its requirements using that contract vehicle, but that the small business portion of that contract vehicle became unavailable after it prepared the IGE.  Supp. Agency Report at 6.  This does not explain why, when it became apparent that it would not be using the Alliant II contract vehicle, the agency did not prepare a new government estimate that took cognizance of the different contract vehicle--the HUBZone small business FSS--that it ultimately used.

There also is no data included with the IGE to show that the labor categories (and, correspondingly, the hourly rates) actually used by the agency in preparing the IGE were appropriate for, and corresponded to, the labor categories that might be necessary to perform the solicited requirement.  The IGE is a 1-page summary document that includes only the most basic calculations used to arrive at yearly total prices for performance of the requirement.[5]  AR, exh. 32, IGE.  There is no supporting data to demonstrate that the labor categories and hourly rates used to perform the calculations were appropriate or reasonable for the solicited labor categories.[6] 

Finally, there is no supporting data to show that the level of effort or labor mix used by the agency in preparing the IGE was appropriate for the solicited requirement.  The agency represents that it used historical data in preparing the IGE, but none of that data is included with the IGE or provided elsewhere in the agency report.  Given CWS’s assertion that the solicited requirements increase the level of effort called for in comparison to the services provided under its predecessor contract, and increase the credentials required for employees to work on the contract, there is no information in the record to show that the agency’s estimated level of effort or skill mix took these considerations into account or were appropriate or reasonable under the circumstances. 

In sum, the record shows that the IGE was prepared using averaged hourly rates derived from a contractual vehicle that is not an appropriate indicator of the potential hourly rates that might be used by the firms here to prepare their quotations; nothing in the record shows that the hourly rates actually used were appropriate for the actual services to be performed; and nothing in the record shows that the level of effort or labor mix used by the agency was appropriate for the solicited requirements.  In light of these considerations, we conclude that the IGE was a largely flawed figure to show the likely cost of the solicited requirement.

The record also shows that the agency determined that CWS’s proposed price was unreasonable based solely on a comparison of that price to the flawed IGE without considering the results of the competition.  The agency defends its actions by claiming that, because this is an FSS acquisition under Federal Acquisition Regulation (FAR) subpart 8.4, it was not required to use any of the price reasonableness analysis techniques enumerated elsewhere in the FAR.[7] 

While the agency may be correct that FAR subpart 8.4 does not require the use of any of the price analysis techniques enumerated in FAR part 15, the agency’s chosen method must be rational.  A comparison of the prices received to one another--and to the IGE--shows that there may have been a problem with the accuracy of the IGE; only one firm (Intellect) submitted a total price below the government estimate.  The remaining three firms all submitted prices higher than the government estimate, and two of those firms (including CWS) submitted prices significantly higher than the IGE.  AR, exh. 33, Price Analysis Excel Workbook, Initial Analysis Spreadsheet.  This spectrum of pricing suggests that there may have been a problem with the accuracy of the IGE and, as discussed above, the record shows that there were several reasons why the IGE may have been a misleading indicator of the likely cost of acquiring the solicited services.  While the agency was not required to use any of the methods for evaluating reasonableness enumerated in FAR part 15, the agency was nonetheless obligated to make reasonable assessments, and the techniques outlined in FAR part 15 provide a useful framework for reviewing the agency’s assessments. 

As a final point of comparison, the protester represents--and the agency has not demonstrated otherwise--that the average hourly rate of compensation under its predecessor contract for all employees is $38.28.  Letter of Protest, at 18; Protester’s Comments at 13.  The IGE assumes an average hourly rate of compensation for all employees of just $33.80 in the base year, $35.49 in the second year of performance, and $37.26 in the third year of contract performance.  AR, exh. 33, Price Analysis Excel Workbook, CSD IGE Spreadsheet.  It is not until the fourth year of contract performance that the average hourly rate of compensation assumed for all employees under the IGE rises above the average level of compensation for all employees under the current contract, to $39.13.  Id.  Further, the IGE assumes an overall average hourly rate of compensation for all employees over the life of the contract of $38.15, which is below the average rate currently being paid under the predecessor contract ($38.28).  This is notwithstanding the fact that the agency used a 5 percent year-over-year escalation factor to account for increases in the cost of living in preparing the IGE.  AR, exh. 27, Price Analysis Report, at 1.  This provides yet another indicator that the IGE was materially flawed and, by extension, that the agency’s determination that CWS’s price was unreasonably high also was materially flawed.  We therefore sustain this aspect of CWS’s protest. 

Determination That One of CWS’s Proposed Labor Rates Was Unrealistic

The agency also disqualified the CWS quotation from further consideration because one of its proposed unburdened hourly rates was determined to be unrealistically low.  In making its realism determinations, the agency compared the firms’ proposed unburdened hourly rates for employees in Washington, D.C. and St. Louis, Missouri, to benchmark unburdened hourly rates the agency identified using the commercial service salary.com for specific labor categories in those geographic locations.[8]  In every instance where a firm’s unburdened hourly rate for a particular labor category was lower than the benchmark rate identified by the agency in salary.com, the proposed rate was automatically deemed unrealistic.  Further, a firm’s quotation was deemed automatically ineligible for award where any one of the evaluated labor rates was found to be unrealistic. 

The agency found that CWS’s quoted unburdened hourly rate for the project manager-senior labor category was unrealistically low because its proposed hourly rate was $59.62, while the benchmark rate the agency deemed realistic from salary.com was $60.00.  AR, exh. 27, Price Evaluation Report, at 13. 

CWS argues that it was irrational for the agency to have found this labor rate unrealistic, and to have disqualified its proposal from further consideration.  According to CWS, the rate at issue is for a single, part-time employee--a senior project manager--who will only be billed to the task order for 160 hours in the base period, and only 120 hours per year thereafter.  CWS maintains that it was unreasonable for the agency to have deemed its entire quotation unrealistic based on the agency’s finding relating to this one employee’s hourly rate of compensation.  CWS also points out that, for the employee in question, its quotation included a letter of intent demonstrating that, in fact, it is able to hire the employee at the hourly rate quoted.  More generally, CWS argues that the agency apparently selected particular rates from salary.com, but nothing in the record shows how those rates were selected.  For example, CWS points out that entering the job category “project manager-senior” in the salary.com website generates over 1,000 possible job classifications, and there is no way to tell which entry was selected by the agency, or why the particular classifications for each job title may have been selected. 

We sustain this aspect of CWS’s protest.[9]  In the context of a fixed-price contract or task order, an agency may evaluate prices for realism for the limited purpose of measuring the vendors’ understanding of the requirements, or to assess the risk inherent in a firm’s proposed approach.  SKE Italy Sri, B-414884.3, Jan. 24, 2018, 2018 CPD ¶ 37 at 6.  Here, the RFQ advised firms that the agency would evaluate unburdened hourly rates for realism as follows:

The Government will review the unburdened hourly labor rates for realism to determine whether the quoted unburdened hourly labor rates are consistent with current market rates, respective to the geographical location of performance: St. Louis and Washington, DC.  Quotations can be rejected for offering unburdened hourly rates so low that they reflect a lack of technical understanding or would impact recruitment and retention of contractor personnel to perform the work.  The Government may use government civilian rates, similar contract labor rates, www.salary.com, other similar or relevant websites, or anything else it deems appropriate or beneficial in its evaluation of realism.  To be eligible for award, the unburdened hourly labor rates must be found to be realistic.

AR, exh. 1, RFQ, at 12.  We find no reasonable basis for the agency’s conclusion that the CWS quotation was unrealistic.  First, the agency made a finding about the realism of the firm’s quotation as a whole based on its evaluation of just one unburdened rate proposed by CWS for a part-time employee.  AR, exh. 27, Price Analysis Report, at 12-13; exh. 25, SSDD, at 4.  Nothing in the record shows that the agency evaluated the realism of the CWS quotation overall, and there is no evidence to suggest that the agency considered CWS’s remaining proposed unburdened rates unrealistic, such that a finding that its quotation overall was unrealistic.  Even if we were to conclude that the agency’s identification of this one labor rate as unrealistic was reasonable (a question discussed further below), there is no basis for the agency to have disqualified the CWS quotation as a whole for offering that labor rate for an employee that will provide less than one tenth of a full-time equivalent (FTE) annually.  Such a de minimis consideration could not reasonably lead the agency to find any appreciable risk of CWS being unable to recruit and retain personnel for the requirement.

More to the point, there was no rational basis for the agency to have concluded--based on the contents of the CWS quotation--that CWS would experience difficulty recruiting or retaining the particular employee whose unburdened hourly rate was found unrealistic.  That is because the CWS quotation included a letter of intent executed by the individual in question.  AR, exh. 17, CWS Revised Technical Proposal, at 17. 

We also agree with CWS that the record essentially is devoid of any information that would show that the benchmark unburdened hourly rates used by the agency for purposes of performing its realism evaluation were rationally selected.  The only information in the record relating to how the benchmark unburdened rates were selected by the agency from salary.com provides as follows:

The Government used comparisons to unburdened hourly labor rates to those found on salary.com in its evaluation of realism.  The labor categories found on salary.com were the same as those proposed by the Offerors so there was no discrepancy in its comparison.  In conducting the evaluation, it was determined that any Offeror’s proposed unburdened rate that was below the salary.com rate was unrealistic as the low rate could impact recruitment and retention and cause high turnover negatively impacting performance.

AR, exh. 27, Price Analysis Report, at 3 (emphasis supplied); see also, id. at 6-7, 9, 12-14. 

Thus, the limited information in the record shows that the agency selected unburdened hourly rates from salary.com based on the labor category nomenclature used by each firm in its quotation.  However, the firms (and the agency in preparing the IGE) each used different labor category nomenclature in many instances.  Compare AR, exh. 23, IS Revised Price Workbook, “Unburdened Labor Categories” Worksheet; AR, exh. 19, CWS Revised Price Workbook, “Unburdened Labor Categories” Worksheet; with AR, exh. 32, IGE.  Accordingly, the record shows that the agency used different benchmark unburdened hourly rates for each quotation based on the labor category nomenclature used by the firms in preparing their respective quotations.  Such a comparison was unreasonable to the extent the firms were not evaluated for realism on a common basis, since it is axiomatic that proposals or quotations must be evaluated on a common basis.  Safal Partners, Inc., B-416937, B-416937.2, Jan. 15, 2019, 2019 CPD ¶ 20 at 4.

In addition, as correctly noted by CWS, using any particular job title or nomenclature on the salary.com website returns multiple listings for that job title or nomenclature.  There is nothing in the record to show that the agency selected unburdened rates that corresponded to the actual positions to be filled in terms of the necessary qualifications.

Finally, a review of the IGE in comparison to the unburdened hourly rates selected by the agency for purposes of performing its realism evaluation shows that, in at least some instances, the fully burdened hourly rates used by the agency to prepare the IGE would have been found unrealistic compared to the unburdened rates used by the agency to evaluate realism.[10] 

For example, for the labor category senior computer user support specialist, the agency used a fully burdened hourly rate of $41.39 to prepare the IGE.  AR, exh. 32, IGE.  In evaluating the realism of the CWS unburdened hourly rates, the agency used a benchmark hourly rate of $45 for Washington, D.C. for the same labor category.  AR, exh. 27, Price Analysis Report, at 6-7.  Thus, using the agency’s own test for evaluating realism, the fully burdened hourly rate used to prepare the IGE for this labor category in Washington D.C. would have been found unrealistic in comparison to the unburdened benchmark rate used to evaluate realism.[11] 

As a second example, the record shows that, for the labor category junior computer user support specialist, the IGE used a fully burdened hourly rate of $26.22.  AR, exh. 32, IGE.  In evaluating the realism of CWS’s unburdened hourly rate for the same labor category, the agency used a benchmark hourly rate of $28 for Washington, D.C.  AR, exh. 27, Price Analysis Report, at 6-7.  Once again, using the agency’s own test for evaluating realism, the fully burdened rate used to prepare the IGE would have been found unrealistic in comparison to the benchmark unburdened rate used to evaluate realism.[12] 

Of particular note, the two labor categories discussed in the above examples comprise the majority of the labor that the agency thought would be necessary to perform the overall requirement.  The record shows that, in preparing the IGE, the agency assumed that it would be necessary to use 136 FTEs to perform the requirement annually.  Based on the hours used to prepare the IGE, the two unrealistic labor categories discussed above comprise a significant majority of the labor--92 of the 136 FTEs--used to calculate the IGE.  This is in contrast to the agency’s finding that the CWS quotation as a whole was unrealistic based on a finding that one of its proposed labor rates was unrealistic, but only for a labor category contributing just one tenth of a single FTE annually to performance.

In light of the discussion above, it appears, based on the limited information available in the record, that:  (1) there was no reasonable basis for the agency to have disqualified the CWS quotation as unrealistic because the quotation, read as a whole, shows that there was no risk that CWS’s unburdened hourly rates were so low as to create a risk that the firm would be unable to recruit or retain personnel to perform the requirement; (2) the agency’s selection of the benchmark labor rates to evaluate realism was largely irrational and may also have resulted in the quotations being evaluated on other than a common basis; and (3)  an overwhelming majority of the FTEs used to calculate the IGE were priced using hourly rates that would have been deemed unrealistic by the agency’s own evaluation methodology.  This final conclusion highlights both the irrationality of the agency’s method for evaluating the realism of the quotations, and also further highlights the fundamental inaccuracy and unreliability of the IGE. 

Evaluation of the Intellect Quotation

CWS also challenges the agency’s evaluation of the Intellect quotation.  CWS, as the incumbent, argues that Intellect was able to achieve its comparatively low price because it offered a level of effort substantially below the level of effort necessary to perform the requirement.  CWS maintains that the agency failed to meaningfully evaluate the comparatively low level of effort offered by Intellect.

We sustain this aspect of CWS’s protest.  The RFQ was explicit concerning what the agency was required to consider in connection with the technical capability evaluation of the quotations in terms of level of effort.  The RFQ provided as follows:  “The Government will evaluate whether the level of effort and the mix of labor quoted is sufficient to meet PWS requirements.”  AR, exh. 1, RFQ at 11. 

Here, as noted by the protester, the record shows that Intellect offered to perform with substantially fewer FTEs than the agency itself thought were necessary to perform the requirements.  As noted above, in preparing the IGE, the agency determined that performance of the task order overall would require 136 FTEs annually.  AR, exh. 32, IGE; exh. 33, Price Analysis Abstract Workbook, CSD IGE Worksheet.  In comparison, the record shows that Intellect offered to perform using only 114 FTEs during the base period and first option year of performance, but this number declined during the out-years of the task order, with Intellect offering just 110 FTEs in the second option year of the contract and only 106 FTEs during the third and fourth option years of performance.  AR, exh. 22, Intellect Revised Technical Quotation at 12; exh. 23, Intellect Revised Price Workbook, Labor Rates Worksheet; exh. 33, Price Analysis Report Workbook, Labor Hours Worksheet.  In effect, therefore, Intellect offered to perform the requirement using up to 30 fewer FTEs (approximately 22 percent, or almost one quarter fewer FTEs) compared to the 136 FTEs the agency thought would be necessary to perform the requirement.

As noted above, the agency assigned the Intellect proposal a rating of high confidence under the technical capability evaluation factor.  However, while the record shows that the agency apparently performed calculations that revealed the steadily-declining level of effort being offered by Intellect, AR, exh. 33, Price Analysis Report Workbook, Labor Hours Worksheet, the record is devoid of any contemporaneous, critical analysis on the part of the agency to evaluate Intellect’s ability to perform the requirement using this steadily-diminishing level of effort. 

The agency claims that, in fact, its technical evaluation clearly took these considerations into account.  But there are only two minimal references to this staffing reduction.  The first appears on page 1 of the agency’s technical evaluation report, where the agency makes a reference to Intellect having provided examples of how it would leverage best practices to create efficiencies and cost savings.  AR, exh. 26, Technical Evaluation Report, at 1.  However, beyond simply mentioning efficiencies and cost savings, the agency‘s evaluation materials do not actually describe these efficiencies or explain how such efficiencies might generate savings.  The second reference was made in connection with the agency’s evaluation of Intellect’s organizational structure.  This second reference provides, in its entirety, as follows:  “The staffing plan discussed initiatives to reduce trouble calls and handle calls more efficiently to provide an opportunity to reduce staff and reduce the cost to the government in option years 2 & 3 which exceeds the PWS requirement.”  Id. at 2.  This is the entirety of the contemporaneous record. 

While the agency claims in response to the protest that it performed a critical analysis of Intellect’s pronounced diminishing level of effort, this analysis is found nowhere in the contemporaneous evaluation record.[13]  As noted above, the RFQ expressly required the agency to critically evaluate the level of effort offered to perform the requirement.  The record does not reflect any meaningful evaluation of Intellect’s steadily-diminishing level of effort, notwithstanding the requirements of the RFQ.  Accordingly, we sustain this aspect of CWS’s protest.[14]

RECOMMENDATION

In light of the foregoing discussion, we sustain CWS’s protest.  We recommend that the agency devise methodologies for evaluating the realism of the firms’ proposed unburdened labor rates, and the reasonableness of their total prices, that are supported by contemporaneous information showing that those methodologies accurately reflect the agency’s requirements, and provide a common and rational basis for the evaluation of quotations.[15] 

Once the agency has taken these steps, we recommend that the agency reevaluate quotations and make a new source selection decision.  Should the agency conclude that a firm other than Intellect is properly in line for award, we recommend that the agency terminate the task order issued to Intellect for the convenience of the government, and issue a task order to the successful firm, if otherwise proper.  Finally, we recommend that the agency reimburse CWS for the costs associated with filing and pursuing its protest, including reasonable attorneys’ fees.  CWS’s certified claim for costs, detailing the time expended and costs incurred, must be submitted to the agency within 60 days after this decision.  4 C.F.R. § 21.8(f)(1).

The protest is sustained.

Thomas H. Armstrong
General Counsel

 

[1] The agency assigned ratings of high confidence, some confidence or low confidence, to quotations under the non-price factors.  AR, exh. 24, Source Selection Plan, at 1-2.

[2] Intellect argues that any challenge to the IGE’s accuracy is untimely because CWS was advised during discussions that its price was deemed unreasonably high in relation to the IGE.  Intellect misunderstands our timeliness rules.  Although the firm is correct that CWS was generally advised during discussions that its price was unreasonably high in relation to the IGE, it was not provided a copy of the IGE or otherwise advised of the methodology or data used to prepare the IGE.  Information allowing CWS to deduce the gross amount of the IGE was provided to CWS for the first time in the letter it received advising that it had not been selected for issuance of the task order.  CWS filed its initial protest challenging the accuracy of the IGE within 10 days of receiving that information.  CWS further refined its challenge to the IGE within 10 days of receiving the agency report, which provided additional details regarding the preparation of the IGE.  We therefore conclude that CWS’s challenge to the accuracy of the IGE was timely filed in accordance with our Bid Protest Regulations, 4 C.F.R. § 21.2(a)(2). 

[3] The GSA Alliant II contract vehicle is a multiple-award, indefinite-delivery, indefinite-quantity contracting program available to all government agencies to fulfill information technology requirements.  See https://www.gsa.gov/technology/technology-purchasing-programs/governmen… (last visited Feb. 25, 2020).

[4] In responding to this allegation, the agency affirmatively represented that it “never claimed” to have used the small business Alliant II hourly rates in developing the IGE, and used what it has described only generically as “the Alliant II rates.”  Supp. Agency Report at 5-6.  We take from the agency’s statements that--apparently--it used averaged hourly rates of large business Alliant II contractors to develop the IGE.

[5] For each year of contract performance, the IGE includes a list of 12 generic labor categories, the number of hours the rates were multiplied by in each labor category, and the extended sum of these calculations.  The extended sums are then added (along with minor amounts for “surge” travel and the transition in/out costs) to arrive at a total calculation for each contract year.

[6] We discuss below apparent realism problems associated with the IGE’s labor rates.

[7] Subsection 15.404-1(b) of the FAR includes an enumerated list of proposal analysis techniques to be used to determine price reasonableness, and this list includes comparison of the prices received to one another.  The FAR also denotes comparison of the prices received, along with comparison of the prices received to historical prices paid for the same or similar items, as the preferred methods for evaluating price reasonableness.  FAR § 15.404-1(b)(3).

[8] These are the principal locations for performance of the task order.

[9] As with CWS’s challenge to the propriety of the IGE, Intellect argues that CWS’s challenge to the agency’s use of the salary.com unburdened hourly rates is untimely because the RFQ announced that this was one possible metric the agency would use for assessing realism of the proposed unburdened rates, and because the agency advised CWS during discussions that it was using salary.com to evaluate realism. 

Intellect is correct that the RFQ advised firms that salary.com was a possible source for data that could be used to evaluate realism, and that CWS was advised during discussions that the agency was using salary.com as a point of reference to evaluate realism.  However, CWS did not know until it was provided notice that the task order had been issued to Intellect that its hourly rate for the program manager-senior labor category was found unrealistic using only salary.com data, or that the benchmark rate used for evaluation purposes was only $0.38 higher than CWS’s proposed hourly rate.  AR, exh. 34, CWS Notice to Unsuccessful Firm, Price Analysis Report.  Prior to that time, CWS did not have information to understand adequately the agency’s basis for evaluating realism.  CWS did not know the actual dollar amount of the salary.com rate used by the agency for comparison purposes, or the labor category identified as unrealistic.  Once CWS learned these details, it filed its protest within 10 days of being provided this information; its protest therefore is timely.  4 C.F.R. § 21.2(a)(2).

[10] Because of a lack of information in the record, we are unable to determine what unburdened hourly rates the agency used to prepare the IGE.  As noted, the IGE only includes a list of labor categories, the number of hours for each labor category, and an extended total price for each labor category.  There is no information in the IGE relating to the indirect rates the agency used to prepare the extended prices.  Had we been able to examine the unburdened hourly rates used by the agency in preparing the IGE, it is entirely possible that more of the hourly rates used by the agency to prepare the IGE--other than the ones discussed in the examples below--also would fail the agency’s test for realism.

[11] To assess realism for the rates proposed in St. Louis for this labor category, the agency used a benchmark hourly rate of $39.  The rate used to calculate the IGE ($41.39) is above the benchmark rate used by the agency to evaluate realism, and therefore would appear realistic.  However, the rate used to prepare the IGE is a fully burdened hourly rate, rather than an unburdened hourly rate; it is therefore possible that the agency’s unburdened rate would fail the agency’s test for realism for employees in the St. Louis area as well. 

[12] The agency used a benchmark unburdened hourly rate of $24 for St. Louis which, once again, appears to show that the hourly rate used to prepare the IGE ($26.22) was realistic.  However, the rate used in the IGE is a fully burdened hourly rate; once again,  it is therefore possible that the agency’s unburdened rate would fail the agency’s test for realism for employees in the St. Louis area as well.

[13] The agency did submit an affidavit from the chairman of its technical evaluation team in which he claimed that, in fact, the evaluators did critically analyze Intellect’s staffing approach.  However, none of that explanation appears anywhere in the contemporaneous record. 

[14] CWS advanced two other protest allegations relating to the agency’s evaluation of the Intellect quotation.  First, CWS argues that the agency did not critically evaluate or analyze the labor mix offered by Intellect; CWS maintains that Intellect was able to achieve additional cost savings by proposing comparatively less-qualified personnel. 

For the same reasons outlined above concerning the agency’s evaluation of Intellect’s proposed level of effort--namely that there is virtually nothing in the contemporaneous record reflecting the agency’s critical analysis of Intellect’s labor mix--we also find that the agency’s evaluation failed to meaningfully evaluate this aspect of the Intellect quotation.

Second, CWS argues that one of Intellect’s corporate experience examples did not meet the RFQ’s requirements relating to the number of users supported.  We need not discuss this aspect of CWS’s protest in detail, since we recommend that the agency perform a reevaluation of quotations.  In performing that reevaluation, the agency may also wish to validate its evaluation finding relating to Intellect’s corporate experience.

[15] In the alternative, the agency may elect to revise its acquisition strategy to change how it intends to evaluate quotations. 

Downloads

GAO Contacts

Office of Public Affairs