THIS OPINION WAS INITIALLY ISSUED UNDER PROTECTIVE ORDER AND IS BEING RELEASED TO THE PUBLIC IN REDACTED FORM ON JULY 18, 1995 _______________________ GRANTED: June 22, 1995 _______________________ GSBCA 13247-P UNISYS CORPORATION, Protester, v. NATIONAL AERONAUTICS AND SPACE ADMINISTRATION, Respondent, and JOHNSON CONTROLS ASSURANCE COMPANY, Intervenor. Richard J. Webber and William W. Goodrich, Jr. of Arent Fox Kintner Plotkin & Kahn, Washington, DC; and William A. Wotherspoon and Anne H. Warner of Unisys Corporation, McLean, VA, counsel for Protester. Vincent A. Salgado, Office of the General Counsel, National Aeronautics and Space Administration, Washington, DC; and Allison K. Lawrence, Office of the Chief Counsel, Goddard Space Flight Center, Greenbelt, MD, counsel for Respondent. Kenneth S. Kramer, Bruce J. Casino, Douglas E. Perry, and Anne B. Perry of Fried, Frank, Harris, Shriver & Jacobson, Washington, DC, counsel for Intervenor. Before Board Judges WILLIAMS, DeGRAFF, and GOODMAN. DeGRAFF, Board Judge. On April 21, 1995, Unisys Corporation (Unisys) filed this protest challenging the decision of the National Aeronautics and Space Administration (NASA) to select Johnson Controls Assurance Company (JCAC) for negotiations leading to award of a contract. Because NASA did not comply with the terms of the solicitation when it made its decision, we grant the protest. Findings of Fact The Solicitation For approximately thirty years, Unisys has been the incumbent contractor providing quality assurance and engineering support services to NASA for the Goddard Space Flight Center. Protest File, Exhibit 18B at 1. On August 19, 1994, NASA issued solicitation RFP5-00300/068 for the succeeding contract to provide these services. The contract's basic performance period is to be two years, with one three-year option period. The contract is to be a cost-plus-award fee type contract. A separate two-month contract will be negotiated for the phase-in costs of a new contractor if Unisys is not the awardee. Protest File, Exhibit 6 at 1, 7, 8, 20, 44-45, 243, 280. The solicitation does not require offerors to provide a certain number of employees to accomplish the work required by the contract. Instead, NASA anticipates awarding a "mission type" contract and the solicitation explains that the contractor will determine the number of employees needed to perform the contract work. Protest File, Exhibit 6 at cover letter, 49; Transcript at 145-46, 583. The solicitation provides that the contract will incorporate a ceiling rate for indirect costs and a ceiling for employee- related phase-out costs. The phase-out costs ceiling will apply if the successful offeror in this procurement is not awarded the next succeeding contract for quality assurance and engineering support services. Protest File, Exhibit 6 at 9. Section L.9 of the solicitation explains that NASA will award a contract to the responsible offeror whose offer "will be most advantageous to the Government, cost or price and other factors, specified elsewhere in this solicitation, considered." Protest File, Exhibit 6 at 250. Section M of the solicitation informs offerors that NASA will take four factors into account when evaluating proposals. The factors are Mission Suitability, Cost/Price, Relevant Experience and Past Performance, and Other Considerations (such as financial condition and labor relations). Section M explains that the Mission Suitability factor indicates "the merit or excellence of the work to be performed." The Cost/Price factor "is used to assess what each offeror's proposal will probably cost the Government" and proposed costs "are analyzed to determine the probable 'cost of doing business' based upon the offeror's proposed approach." Protest File, Exhibit 6 at 321. Mission Suitability and Cost/Price are of "essentially equal" importance and are more important than the other two factors. Protest File, Exhibit 6 at 322. The Mission Suitability factor is the only evaluation factor to be point scored. The available points, which reflect the relative importance of the Mission Suitability subfactors and elements, are: Understanding of Requirements 450 Technical approach 350 Professional employees compensation plan 100 Personnel, Organization, and Staffing Plan 400 Key personnel 175 Organization and staffing plan 225 Company Resources 150 Total Points 1000 Protest File, Exhibit 6 at 324-25. Section M explains that cost/price will be evaluated to establish "[t]he realism of the proposed cost/price," "[t]he probable cost to the Government" and, "[i]f ceiling limitations are involved, the maximum cost to the Government for the proposed effort." Protest File, Exhibit 6 at 328-29. Unisys and JCAC submitted proposals in response to the solicitation. Protest File, Exhibits 17, 18. JCAC is a newly- formed joint venture between Johnson Controls World Services, Inc. (Johnson Controls) and System Planning Corporation (SPC). Protest File, Exhibit 17 at 1. The Evaluation NASA appointed a business evaluation panel (BEP) and a technical evaluation panel (TEP) to review the offerors' proposals. Protest File, Exhibit 2. NASA decided that both Unisys and JCAC had a reasonable chance of being selected for award and held discussions with both offerors. Protest File, Exhibits 19, 20, 21. After discussions, in early February 1995, Unisys and JCAC submitted their best and final offers. Protest File, Exhibits 23, 24. The Business Evaluation Panel The BEP evaluated the Other Considerations factor and assigned Unisys a rating of "excellent" and JCAC a rating of "very good." The BEP's findings concerning the Relevant Experience and Past Performance factor were combined with the findings of the TEP, discussed below. Protest File, Exhibit 25 at 7, 10; Transcript at 616-17. The BEP also evaluated the Cost/Price factor. The BEP concluded that, for the contract's basic two-year period plus the three-year option period, Unisys' probable cost plus maximum award fee will be ************, and JCAC's probable cost plus maximum award fee will be ***********. These amounts include overhead calculated at Unisys' and JCAC's proposed ceiling rates and do not include phase-out costs. Protest File, Exhibits 25, 27. Several aspects of the BEP's evaluation of the Cost/Price factor are at issue in this protest. They are: Vacancy credit An employee of SPC who was responsible for preparing JCAC's proposal explained that, during SPC's performance of a contract at Goddard Space Flight Center, an average of ******-*** hours per year per position were not billed to the customer. These hours are attributable to unpaid leave and to the time that a position is vacant after an employee leaves and before the employee is replaced. Transcript at 683-85. In its proposal, JCAC includes a credit based upon ******-*** hours per year per position not being worked or billed. Protest File, Exhibit 23C at 8-9, 19. The SPC employee stated that the vacancy credit is based upon his experience with a number of contracts. Transcript at 710. The SPC employee testified that SPC and Johnson Controls include a vacancy factor in virtually all bids for support services contracts. Transcript at 690. Not all employees will take ******-*** hours per year of uncompensated leave and not all employees will be replaced every year. Transcript at 704. The BEP did not eliminate the vacancy credit when it arrived at JCAC's probable cost. Protest File, Exhibit 27. Vacation JCAC proposes *** ***** of paid vacation and two weeks of paid holidays for personnel subject to the Service Contract Act. Protest File, Exhibit 23C at 8. The Department of Labor wage determination contained in the solicitation explains that personnel subject to the Service Contract Act are to be provided with two weeks of paid holidays and with two, three, or four weeks of paid vacation, depending upon length of service. Length of service includes time spent working for Unisys on the support services contract. Protest File, Exhibit 6 at 217. JCAC's proposal states that JCAC employees will earn vacation time in accordance with the Department of Labor wage determination and acknowledges that JCAC will abide by the Service Contract Act requirements. Protest File, Exhibit 23C at 8-9. JCAC proposes **** ***** of paid vacation against which holidays will be charged for personnel not subject to the Service Contract Act. Protest File, Exhibit 23C at 8. JCAC estimates that it will hire ****** percent of the current Unisys employees. Protest File, Exhibit 17D at 175; Transcript at 710. Unisys includes *** *** ***-**** ***** of paid vacation for all employees in its proposal. Protest File, Exhibit 18B at 47. The BEP did not make any adjustments related to vacation when it arrived at JCAC's probable cost. Protest File, Exhibit 27. Cost of Three Employees In its initial proposal, JCAC showed that it intended to fill three specified positions with full-time employees and JCAC included the cost of these employees in its proposal. Protest File, Exhibit 17A at 70. In its best and final offer, JCAC eliminated these employees and reduced its cost accordingly. Protest File, Exhibit 23C at 32, 37. The SPC employee testified that the functions which were to have been performed by one of these employees will be performed by employees of either Johnson Controls or SPC and that the costs of these Johnson Controls and SPC employees are included in JCAC's general and administrative expense pool. The SPC employee stated that, in its best and final offer, JCAC increased its general and administrative ceiling rate and that the increase was, in part, to account for the added cost associated with these employees. The SPC employee also testified that the functions which were to have been performed by two of the three employees will be performed by JCAC employees and that the costs of these employees are included in JCAC's direct labor costs and in its general and administrative expense pool. Transcript at 697-701. In its best and final offer, in response to questions from NASA, JCAC stated that ** ********* *** ******* *** ************** ******* **** *** ******* * ***** ****** ** ***** *** ************* ***** **** ***** ******* Protest File, Exhibit 23B at 28. The BEP did not make any adjustments related to these three positions when it arrived at JCAC's probable cost. Protest File, Exhibit 27. The TEP did not make any adjustments in JCAC's staffing plan related to these three positions. Protest File, Exhibit 26. Full-time employees The chair of the BEP expected that, when responding to a solicitation for a mission type contract, offerors would propose different employee skill mixes and different labor categories. In fact, Unisys and JCAC proposed different skill mixes and labor categories. Transcript at 584-85. The SSO testified that the staffing levels for both contractors will be roughly the same. Transcript at 145. JCAC proposed the equivalent of ***** full-time employees and Unisys proposed the equivalent of *** full-time employees. Protest File, Exhibit 26, Enclosure 3 at 2, 3. The TEP reviewed each offeror's proposed staffing levels and determined that JCAC's proposal needed an ********** **** full-time employees in order to reflect more accurately the personnel needed to perform some of the tasks required by the contract. Protest File, Exhibit 26, Enclosure 3. The BEP's evaluation of JCAC's probable cost includes ** ****** adjustment to account for an ********** **** full-time employees, for a total of ***** full-time employees.[foot #] 1 Protest File, Exhibit 25. Key personnel In its best and final offer, JCAC listed five key personnel. JCAC's cost proposal is based upon the assumption that *** ** ***** *** ********* **** ***** **** ***** *** ***-**** ***** **** ** *** ******** *** **** ** ******** ** *********** *** ******* ***** ********. Protest File, Exhibit 23C at 30. In its best and final offer, JCAC stated that it assumed some employee turnover would be part of any contract of this size and duration. JCAC also stated that, when ******** **** ******-*********** ********* ***** *** ********* **** **** **** ** ******* ***** *********** **** *****-********** *** **** ****** *********** JCAC stated that it used wage surveys in order to estimate the cost of replacement personnel. Protest File, Exhibit 23C at 8. The BEP did not make any adjustments related to key personnel when it arrived at JCAC's probable cost. Protest File, Exhibit 27. Depreciation - Overhead Ceiling The BEP calculated each offeror's probable cost by using the offeror's ceiling overhead rates.[foot #] 2 Protest File, Exhibits 25, 27. The BEP's evaluation of Unisys' probable cost includes an upward adjustment to Unisys' proposed ceiling overhead rate. The BEP made this adjustment because the BEP determined that the solicitation requires offerors to include $200,000 per year for equipment depreciation in their overhead cost pools and Unisys included less than this amount. When the BEP included the $200,000 in Unisys' overhead cost pool, this caused Unisys' bid overhead rate to increase. The BEP increased Unisys' proposed ceiling overhead rate by an amount which will maintain the *** ******* difference proposed by Unisys between ----------- FOOTNOTE BEGINS --------- [foot #] 1 Actually, the BEP adjusted the number of full-time equivalents and not the number of full-time employees. The distinction is not material to our decision. [foot #] 2 An offeror expects to incur overhead costs at its bid overhead rate. If an offeror incurs overhead costs at a rate in excess of its bid overhead rate, the offeror can recover its costs up to its ceiling overhead rate. ----------- FOOTNOTE ENDS ----------- its bid overhead rate and its ceiling overhead rate. Protest File, Exhibits 25, 18B at 28; Transcript at 610-11, 655-57. The chair of the BEP testified that, if Unisys had been selected for negotiations leading to award, its contract would not necessarily have contained its proposed ceiling overhead rate. Instead, Unisys' expressed desire to maintain a "spread between bid and ceiling" would have been a subject of negotiations. Transcript at 656. On April 25, 1995, NASA told Unisys about this adjustment and explained the rationale for making the adjustment. Transcript at 611, 656-57, 676. Equipment In its proposal, Unisys states that each year it will "procure ******** worth of hardware, software tools, or training for use in providing [Goddard Space Flight Center] applications." Protest File, Exhibit 18B at 134. Unisys' proposal twice states that it will provide "up to" ******** of hardware and software per year. Protest File, Exhibit 18B at 135. Unisys' proposal also states that, each year, it will provide five days of labor worth ****** to review and assist in system development. Protest File, Exhibit 18B at 135. During discussions, Unisys stated that its proposal should have read, "******** per year non- cumulative." Unisys explained that these costs would not be billed to NASA and that the equipment purchased would be a Unisys asset. Protest File, Exhibit 24B at 9. "Non-cumulative" means that if, for example, Unisys provided only ******* of equipment in one year, Unisys would not be obligated to carry forward the ******* "deficiency" into succeeding years. Transcript at 438. The BEP did not make any adjustments related to equipment when it arrived at Unisys' probable cost. Protest File, Exhibit 27. Probable award fee The BEP's calculation of probable cost assumes that the successful offeror will be paid 100% of its proposed award fee. Protest File, Exhibit 27 at 42, 44. The contracting officer, who is also the chair of the BEP, testified that in her seventeen years of experience in Government procurements with several agencies, she has never known the government to evaluate award fee at less than 100% and that to do so would be "sheer speculation." Transcript at 593. A witness who was called by Unisys to provide expert testimony concerning cost and price analysis stated that, in his opinion, the BEP should have assumed that offerors will be paid something less than 100% of their proposed award fees. Transcript at 406-07. The Unisys expert testified that he is not aware of any other expert who has testified that award fee should be evaluated at less than 100% and that he is not aware of the Government ever evaluating award fee at less than 100%. Transcript at 406-08. The solicitation does not state whether award fee will be evaluated at 100% at something other than 100%. Protest File, Exhibit 6. Phase-out costs NASA's calculation of probable cost assumes that offerors will incur 100% of their proposed phase-out costs at the end of the contract.[foot #] 3 Protest File, Exhibits 18B at 33, 23C at 12, 27 at 42, 44. The chair of the BEP testified that in her seventeen years of experience in government procurements with several agencies, she has never known the Government to evaluate phase-out costs at less than 100%. Transcript at 591- 92. The Unisys expert witness stated that, in his opinion, the BEP should have assumed that offerors have a 50% chance of incurring phase-out costs and should have included only 50% of the offerors' proposed phase-out costs in the calculation of probable cost. Transcript at 478-79. The Unisys expert testified that he is not aware of any other expert who has testified that phase-out costs should be evaluated at less than 100% and that he is not aware of the Government ever evaluating phase-out costs at less than 100%. Transcript at 479. The solicitation does not state whether phase-out costs would be evaluated at 100% or at something other than 100%. Protest File, Exhibit 6. Phase-in costs As mentioned above, the solicitation provides that a separate contract will be negotiated for the phase-in costs of a new support services contractor. These costs are associated with a new contractor replacing Unisys, the incumbent contractor. Unisys will not experience any phase-in costs and so was not required to propose any amount for these costs. Transcript at 587, 589-90. The chair of the BEP testified that to consider phase-in costs in the evaluation of a non-incumbent's probable cost would provide the incumbent contractor with an unfair advantage. Transcript at 590. JCAC proposed ******** for phase- in costs. Protest File, Exhibit 23C at 3, 12. The BEP did not include phase-in costs when it arrived at JCAC's probable cost. Protest File, Exhibit 27. The solicitation does not state whether the cost of the phase-in contract would be included in evaluating the probable cost of the support services contract. Protest File, Exhibit 6. Phase-out of Unisys' current contract Unisys' current support services contract provides that Unisys will be paid for its phase-out costs if Unisys is not awarded the contract that is the subject of this protest. Transcript at 208. These phase-out costs were proposed and evaluated five years ago when the current support services ----------- FOOTNOTE BEGINS --------- [foot #] 3 Actually, the BEP's calculations include more than 100% of JCAC's proposed phase-out costs. The BEP included proposed phase-out costs of ******* for JCAC. This amount is the sum of *******, which is the amount proposed by JCAC, and *******, which is the amount proposed by a JCAC subcontractor. Protest File, Exhibits 27 at 44, 23C at 12, 23D at 9. ----------- FOOTNOTE ENDS ----------- contract was awarded to Unisys. Transcript at 591, 659. The BEP did not include these phase-out costs when it calculated JCAC's probable cost. Protest File, Exhibit 27. The contracting officer, who is also the chair of the BEP, explained that consideration of these costs as part of the probable cost of awarding a contract to someone other than Unisys would be unfair because Unisys would always have an advantage over other offerors. Transcript at 659-60. The Unisys expert witness stated that, in his opinion, the BEP should have included as part of JCAC's probable cost the phase-out costs that Unisys will be paid under its existing contract. Transcript at 414. The Unisys expert testified that he is not aware of any other expert who has testified that phase-out costs should be included in a probable cost analysis and that this was the first time he had given such testimony. Transcript at 416. The solicitation does not state whether the phase-out costs of the current Unisys contract would be included in evaluating the probable cost of the support services contract. Protest File, Exhibit 6. Other JCAC costs The BEP did not make any adjustments to JCAC's proposed cost to account for JCAC's conflict of interest plan or JCAC's inflation rate assumptions.[foot #] 4 Protest File, Exhibit 27. The Unisys expert witness briefly testified that, in his view, NASA is likely to incur some additional costs due to JCAC's implementation of its conflict of interest plan and due to JCAC's inflation rate assumption. Transcript at 373-74. The expert's written report does not mention either of these items. Protest File, Exhibit 136. The basis for the expert's opinion concerning JCAC's conflict of interest plan is not entirely clear, although it seems to be based upon the belief that JCAC will have difficulty complying with the solicitation's conflict of interest provisions. Transcript at 373. The basis for his view concerning inflation is that JCAC's inflation rate estimate is lower than the Unisys estimate. Transcript at 374. JCAC's inflation rate estimate is **** and Unisys' estimate is ****. Protest File, Exhibits 23C at 8, 18B at 25. The Technical Evaluation Panel The TEP evaluated the Relevant Experience and Past Performance factor and assigned adjectival ratings of "excellent" to Unisys and "very good" to JCAC. Protest File, Exhibit 26. NASA contacted fourteen organizations concerning the past performance and experience of Johnson Controls, SPC, and their two subcontractors. Based upon these contacts, the TEP, with ----------- FOOTNOTE BEGINS --------- [foot #] 4 The solicitation requires offerors to avoid conflicts of interest that would arise if, for example, they were to sell hardware to NASA under another contract and then test the performance of that hardware under this contract for support services. Protest File, Exhibit 6. ----------- FOOTNOTE ENDS ----------- input from the BEP, determined that the technical performance of SPC and the two subcontractors was in the "very good" to "excellent" range and that the technical performance of Johnson Controls was in the "good" range. The TEP determined that the ability to meet customers' schedules of Johnson Controls, SPC, and the two subcontractors was in the "very good" to "excellent" range. Finally, the TEP determined that the business/cost performance of Johnson Controls, SPC, and the two subcontractors was in the "very good" to "excellent" range, with the exception of one contract at ******* *** ***** ****. Protest File, Exhibit 26 at 46-48; Transcript at 617-20. The TEP also evaluated the Mission Suitability factor. In arriving at Mission Suitability subfactor and element scores, the TEP looked to see whether the offerors had any major or minor strengths or weaknesses in any subfactors or elements. The TEP determined that Unisys has twelve major strengths, twenty-seven minor strengths, and five minor weaknesses in its technical approach; one major strength and two minor strengths in its professional employees compensation plan; one major strength and one minor strength in key personnel; one minor strength in its organization and staffing plan; and one major strength in its company resources. Protest File, Exhibit 26, Enclosure 2. The TEP chair explained that some of Unisys' major strengths could result in cost savings or in cost avoidance. Transcript at 28- 38. The TEP determined that some of the major strengths in Unisys' proposal have the potential to improve efficiency, reduce risk, and minimize costs. Protest File, Exhibit 26, Enclosure 2 at 3-6. The TEP determined that JCAC has two major strengths, seventeen minor strengths, one major weakness and eight minor weaknesses in its technical approach; one minor weakness in its professional employees compensation plan; one minor strength in key personnel; one minor strength and one minor weakness in its organization and staffing plan; and one major strength in its company resources. Protest File, Exhibit 26, Enclosure 2. The major weakness in JCAC's proposal is the result of its lack of understanding of the solicitation's requirement for software assurance, especially concerning test and verification. This risk remained, even though NASA asked JCAC questions about this area during discussions. Protest File, Exhibit 26, Enclosure 2 at 29. The solicitation requires that the contractor perform software assurance for software development activities for spacecraft and ground data systems projects. One part of this requirement is that the contractor will monitor and evaluate test plans, procedures, and results to ensure that requirements are verified. Protest File, Exhibit 6 at 103-04. The TEP chair explained that software assurance is a part of most of the projects at Goddard Space Flight Center and that the risk associated with this weakness is that a system, such as spacecraft, might not work as intended. Transcript at 40-43. The SSO testified that the source evaluation board told him that this major weakness in JCAC's proposal could be corrected by JCAC. As for how this correction would be made, the SSO explained that, after award, NASA could tell JCAC what was wrong with its approach to the software assurance requirement and could explain to JCAC how to perform this requirement. Transcript at 107-110, 113. The TEP scored the Mission Suitability factor as follows: Unisys Understanding of Requirements *** Technical approach *** Professional employees compensation plan ** Personnel, Organization, and Staffing Plan *** Key personnel *** Organization and staffing plan *** Company Resources *** Total Points - Unisys *** JCAC Understanding of Requirements *** Technical approach *** Professional employees compensation plan ** Personnel, Organization, and Staffing Plan *** Key personnel *** Organization and staffing plan *** Company Resources *** Total Points - JCAC *** Protest File, Exhibit 26. The chair of the TEP explained that an offeror's technical approach score predicts how well the offeror will perform the technical tasks required by the contract. A difference in scores between offerors reflects a difference in performance risk, and increased performance risk can result in higher costs to NASA. Transcript at 18. The TEP chair also testified that the key personnel and the organization and staffing plan scores reflect qualities that can affect risk and cost. Transcript at 19-20. The TEP assigned adjectival ratings to the Mission Suitability factor, subfactors, and elements, according to the numerical scores.[foot #] 5 An "excellent" rating means that the proposal is "comprehensive and thorough," "of exceptional merit," and contains either no weaknesses or minor correctable weaknesses. A "very good" rating means that the proposal "demonstrates overall competence" and contains strengths which outweigh weaknesses and that any major weaknesses contained in the proposal are correctable. A "good" rating means that the proposal contains a "reasonably sound response," that weaknesses do not "significantly detract from the offeror's response," and that major weaknesses are probably correctable. Protest File, Exhibit 27 at 12. ----------- FOOTNOTE BEGINS --------- [foot #] 5 A rating of "excellent" equals a point score of ninety-one percent or more. A "very good" rating equals a point score of seventy-one to ninety percent. A "good" rating equals a point score of fifty-one to seventy percent. 48 CFR 1870.303, ch. 4 (1994). ----------- FOOTNOTE ENDS ----------- The adjectival ratings assigned by the TEP to the Mission Suitability factor, subfactors, and elements are as follows: Unisys Understanding of Requirements Technical approach ********* Professional employees compensation plan **** **** Personnel, Organization, and Staffing Plan Key personnel **** **** Organization and staffing plan **** Company Resources ********* Overall rating - Unisys **** **** JCAC Understanding of Requirements Technical approach **** **** Professional employees compensation plan **** Personnel, Organization, and Staffing Plan Key personnel **** Organization and staffing plan **** Company Resources ********* Overall rating - JCAC **** **** Protest File, Exhibits 26, 27. The TEP chair explained that the offerors' strengths and weaknesses are reflected in their numerical scores and in the adjectival ratings. Transcript at 70. The Selection Decision The source evaluation board (SEB), which was composed of the members of the TEP and the BEP, presented its findings to the SSO on March 8, 1995. The SEB informed the SSO of the BEP's conclusions concerning probable cost and the adjectival ratings for the Other Considerations factor. The SEB informed the SSO of the TEP's adjectival ratings for the Relevant Experience and Past Performance factor and the scores and adjectival ratings for the Mission Suitability factor. The SSO and the SEB discussed the major strengths and weaknesses of each offeror. Protest File, Exhibit 27; Transcript at 137, 625. At the conclusion of the SEB's March 8, 1995 presentation, which lasted approximately three hours, the SSO asked the TEP to quantify the potential cost savings associated with Unisys' technical advantages by assigning a dollar value to those advantages. Transcript at 64, 81-82. The SSO wanted to determine whether he could justify an award to Unisys at its higher price. Transcript at 132. The TEP chair believed that, considering Unisys' major strengths, the potential for cost avoidance and cost savings was real. But, the TEP was not able to decide how to assign dollar values to the strengths. Transcript at 65. On March 16, 1995, the SSO again met with the SEB. Because the TEP was unable to quantify the value of the Unisys technical advantages, the SEB was prepared to provide the SSO with greater detail concerning the strengths and weaknesses of the proposals. The SSO believed that the March 16 presentation was essentially a "rehash" of the March 8 presentation, and so the SEB did not review with the SSO all of the material it had prepared for the March 16 meeting. Transcript at 66-67, 83, 104. The SSO did not attempt to quantify the value of the Unisys technical advantages. Transcript at 135. A briefing chart that the SEB prepared for the SSO states that the difference in cost between the two offerors "is insignificant compared to the potential programmatic effect of technical performance" and that contract performance will affect NASA programs worth approximately $2 billion per year. Protest File, Exhibit 28 at 22. The view expressed in the briefing chart is the same as the view of the chair of the BEP, who will also be the contracting officer for the contract at issue in this protest. Transcript at 644, 648-49. At the end of the March 16, 1995 meeting, the SSO announced that he had selected JCAC for negotiations leading to award of the contract. Transcript at 68. On March 24, 1995, the SSO signed a source selection memorandum. The memorandum summarizes the procurement events and the TEP's and BEP's evaluation of the award factors. The SSO's memorandum states that the SSO carefully reviewed the presentation and comments of the TEP and the BEP, the Mission Suitability point scores and adjectival ratings, the proposed and probable costs, the assessment of the Relevant Experience and Past Performance factor and the Other Considerations factor. The memorandum concludes that JCAC's offer will be more advantageous to the Government and selects JCAC for negotiations leading to award of a contract. The SSO explained his decision as follows: This determination is based on the finding that both proposals were in the "very good" range for Mission Suitability with the JCAC proposal offering the lower cost to the Government. It was determined that the technical advantages of the Unisys proposal were not significant enough to overcome the clear price advantage offered by the JCAC proposal. Protest File, Exhibit 29. At the hearing, the SSO testified that Unisys submitted a better technical proposal, that the difference between the offerors' scores is "substantial" or "measurable," that Unisys "understood the job better" than JCAC, and that Unisys' proposal provides a "superior technical response." Transcript at 88-92, 109-10. The SSO's decision was not affected by the fact that Unisys received higher ratings for the Relevant Experience and Past Performance factor and for the Other Considerations factor, because the SSO did not see that there was any problem with JCAC in either of these factors. Transcript at 173-74. The SSO was not influenced by the difference of *** points in Mission Suitability factor scores, because both offerors received the same adjectival rating. The SSO testified that the adjectives "transcend" the importance of the point scores. Transcript at 109. The SSO testified that, even if Unisys had been awarded a few more points and had been given an "excellent" rating, he would have chosen JCAC for award. Transcript at 110. He stated that he believed that the Unisys technical advantages had to be quantified in order for him to decide that they outweighed the cost difference between the two offers. Transcript at 132. The SSO expected that either JCAC or Unisys could perform the requirements of the contract, that their differences would "wash out over time," and that JCAC's proposal represents a cost savings. Transcript at 112-13, 205-06. The SSO explained that JCAC has a "very good" rating and there is a cost difference between JCAC and Unisys, and this "would tend to indicate" that JCAC is the best value for NASA. Transcript at 96. The SSO testified that there is no reason to believe that JCAC will not become a very good contractor. Transcript at 98-9. In the SSO's view, the fact that he expected JCAC to perform well, added to the fact that JCAC offered the lowest probable cost, made JCAC the best value for NASA. Transcript at 108-09. The SSO explained that it would have been difficult to write a source selection statement choosing Unisys, with its higher cost, for negotiations leading to award because source selection statements refer only to adjectival ratings, not point scores, and both JCAC and Unisys received the same overall adjectival rating for the Mission Suitability factor. Transcript at 136-37, 139-41. The Protest Unisys filed its protest complaint on April 21, 1995, and amended the complaint on May 16 and May 25, 1995. A hearing was held on May 30 through June 1, 1995. The parties filed their post-hearing briefs on June 9, 1995, and their reply briefs on June 13, 1995. Unisys challenges NASA's evaluation of JCAC's experience and past performance (Count IV-A) and NASA's cost/technical tradeoff (Count V). Unisys also challenges the following elements of NASA's evaluation of the Cost/Price factor (Counts III and IV): Vacancy credit: Unisys speculates that the ******-*** hours will be worked, billed, and paid by NASA. Unisys contends that the assumptions which underlie the vacancy credit are just as true for Unisys as for JCAC. Unisys calculates that JCAC's probable cost should be increased by ********** in order to eliminate the effect of the credit included in JCAC's proposal. Protest File, Exhibit 136. Vacation: Unisys contends that some of the employees to be hired by JCAC will be entitled to more paid vacation than the amount included by JCAC in its proposal. Unisys calculates that the difference between the paid vacation time included in JCAC's proposal and the paid vacation time which should have been included in JCAC's proposal amounts to ********. Unisys contends that JCAC's probable cost should be increased by this amount. Protest File, Exhibit 136. Cost of three employees: Unisys calculates the cost of the three employees who were eliminated in JCAC's best and final offer as ******** and contends that JCAC's probable cost should be increased by this amount. Protest File, Exhibit 136. Full-time employees: Unisys contends that there is no reason to suppose that JCAC can accomplish the contract work with fewer employees than can Unisys. Unisys calculates that if JCAC were to use the same number of employees as Unisys proposes to use, JCAC's probable cost would increase by **********. Protest File, Exhibit 136. Key personnel: Unisys contends that NASA's technical evaluators considered the capabilities of the individuals listed by JCAC as its key personnel, and so NASA should have increased JCAC's probable cost by ******** to account for *** ********** ****** **** *** *** *** ********* **** ** **** ** **** ** *** ***** ****** *** ******** ****. Protest File, Exhibit 136. Depreciation - overhead ceiling: Unisys contends that the BEP should not have adjusted its proposed ceiling overhead rate upward after revising the Unisys bid overhead rate to account for annual equipment depreciation of $200,000. According to Unisys' calculations, its probable cost should be decreased by ******** to account for the fact that the BEP adjusted its proposed ceiling overhead rate. Protest File, Exhibit 136. Equipment: Unisys contends that NASA should have adjusted Unisys' probable cost by reducing it by ******** ********* *** **** *** **** ****** to take into account Unisys' commitment to providing equipment and labor. Protest File, Exhibit 136. Probable award fee: Unisys contends that the BEP should have assumed that the successful offeror will be paid ninety percent of its proposed award fee. If the BEP had made this assumption, Unisys' probable cost would have been lowered by ******** and JCAC's probable cost would have been lowered by ********. Protest File, Exhibit 136. Phase-out costs: Unisys contends that the BEP should have included only fifty percent of each offeror's proposed phase-out costs in its calculation of probable cost. If the BEP had included fifty percent of each offeror's phase-out costs in its calculation of probable cost, Unisys asserts that its probable cost would have been lowered by ******** and JCAC's probable cost would have been lowered by *******. Protest File, Exhibit 136.[foot #] 6 Phase-in costs: Unisys contends that NASA should have adjusted JCAC's probable cost upward by ********, which is the amount JCAC proposed for the separate contract for phase-in costs. Protest File, Exhibit 136. Phase-out of Unisys' current contract: Unisys contends that the BEP should have increased JCAC's probable cost by ********, which is the amount of phase-out costs that Unisys asserts it will be paid by NASA pursuant to the terms of the current contract. Protest File, Exhibit 136. Other JCAC costs: Unisys contends that the BEP should have considered the cost of implementing JCAC's conflict of interest plan and the greater inflationary risk of JCAC's proposal. Discussion We deny Counts III, IV, and IV-A because Unisys did not establish that NASA violated any statute, regulation, or term of the solicitation when it evaluated either the Relevant Experience and Past Performance factor or the Cost/Price factor. We grant Count V because Unisys established that NASA's decision to conduct negotiations with JCAC was not made in accordance with the terms of the solicitation. Relevant Experience and Past Performance Factor In Count IV-A of its complaint, Unisys alleges that JCAC's rating of "very good" for the Relevant Experience and Past Performance evaluation factor is not supported by the evidence and that the rating ignores the fact that JCAC is a new joint venture. Unisys alleges that JCAC's rating is based, in part, upon an evaluation of one of JCAC's subcontractors which poses a conflict of interest as set forth in Count VI of the complaint. In its post-hearing briefs, Unisys presents no argument in support of Count IV-A. JCAC's "very good" rating for the Relevant Experience and Past Performance factor is based upon NASA's survey of fourteen customers of the JCAC joint venture partners and JCAC's two ----------- FOOTNOTE BEGINS --------- [foot #] 6 Unisys includes one-half of the amount attributable to JCAC, and not JCAC's subcontractor, in arriving at its downward adjustment. ----------- FOOTNOTE ENDS ----------- subcontractors. The technical performance of SPC and the subcontractors was in the "very good" to "excellent" range and the technical performance of Johnson Controls was "good." The schedule performance of both partners and subcontractors was in the "very good" to "excellent" range. The business/cost performance of both partners and subcontractors was also in the "very good" to "excellent" range, with the exception of one contract. Contrary to Unisys' allegation, this evidence supports the "very good" rating assigned to JCAC by NASA for the Relevant Experience and Past Performance factor. Unisys has not shown that JCAC's status as a newly-formed joint venture should have had any effect upon NASA's rating. Also, Unisys amended its complaint to delete the conflict of interest allegations contained in Count VI and presented no evidence in support of the conflict of interest allegation contained in Count IV-A. Count IV-A is denied because Unisys has not established that NASA violated statute, regulation, or the terms of the solicitation when it rated JCAC as "very good" for the Relevant Experience and Past Performance factor. Evaluation of the Cost/Price Factor NASA and JCAC assert that many of Unisys' challenges to the evaluation of the Cost/Price factor are untimely. We disagree. Considering the merits of Unisys' challenges, we hold that Unisys fails to establish, by a preponderance of the evidence, that NASA's evaluation violates statute, regulation, or the terms of the solicitation. Timeliness Our rules require that a protester who wishes to challenge the terms of a solicitation must file its complaint before the date that initial proposals are due. Rule 5(b)(3)(i). NASA and JCAC contend that Unisys did not file a timely challenge concerning the manner in which the Cost/Price factor evaluation takes into account full-time employees, probable award fee, phase-in costs, phase-out costs, and phase-out costs of Unisys' current contract. They argue that Unisys' challenge concerning these items amounts to a challenge to the terms of the solicitation and is untimely because the protest complaint was filed after initial proposals were submitted. Unisys contends that its challenge is timely because it takes issue with the manner in which the evaluation was performed and not with any solicitation provisions. We agree with Unisys. The solicitation does not state whether or how NASA will take into account differences in the numbers of full-time employees, phase-in costs, or the phase-out costs of Unisys' current contract. The solicitation does not state whether NASA will evaluate probable award fee and phase-out costs at 100%. The terms of the solicitation do not give rise to Unisys' complaint. Instead, Unisys' complaint is based upon the manner in which the BEP performed the evaluation of the Cost/Price factor. Our rules also require that a protester who wishes to challenge something other than the terms of a solicitation must file its complaint within ten days after it knows or should have known of the basis for its ground of protest. Rule 5(b)(3)(ii). NASA and JCAC contend that Unisys did not file a timely challenge concerning the manner in which the Cost/Price factor evaluation takes into account the adjustment to Unisys' proposed ceiling overhead rate. They argue that Unisys learned of the manner in which NASA adjusted Unisys' ceiling rate and the manner in which NASA utilized that ceiling rate on April 25, 1995, and that Unisys did not challenge the adjustment until more than ten days later, when it amended its complaint on May 16, 1995. Unisys contends that its challenge is timely because its April 21, 1995 complaint protests the manner in which NASA utilized Unisys' adjusted ceiling overhead rate. We agree with Unisys. The April 21, 1995 complaint states, "NASA's adjustment of Unisys overhead rates is improper to the extent that NASA is using ceiling rates as the basis for evaluation." Complaint 50 (Apr. 21, 1995). This allegation is sufficient to state Unisys' claim under the principles of notice pleading. Vacancy Credit NASA did not violate statute, regulation, or the terms of the solicitation when it included JCAC's vacancy credit in the calculation of JCAC's probable cost. There is no evidence to support Unisys' theory that the ******-*** hours per year per position for which JCAC included a credit will, in fact, be worked and paid for by NASA. Instead, the evidence establishes that SPC and Johnson Controls include a vacancy credit in many of their proposals for support services contracts and that this credit is based upon past experience, including experience with a contract at Goddard Space Flight Center. The fact that not all positions will be vacant for ******-*** hours per year does not undermine JCAC's reason for applying the credit, because the ******-*** hours is not the number of vacancy hours experienced by each position. Rather, it is an average of the number of vacancy hours experienced. Unisys did not propose a vacancy credit. Yet, Unisys argues that NASA should have included a vacancy credit when it evaluated Unisys' probable cost. Unisys asks us to assume that, if JCAC will experience ******-*** hours per year per position which will not be worked or paid for by NASA, Unisys will have the same experience and, therefore, Unisys should be given a vacancy credit. We are not willing to make the assumption that underlies Unisys' argument. Unisys has been the support services contractor at Goddard Space Flight Center for nearly thirty years. If Unisys has experienced ******-*** non-billable hours per position per year, then Unisys could have included a credit in its cost proposal. We do not know whether Unisys' past experience has been the same as SPC's and Johnson Controls' or whether, due to differences in leave policies and efficiency in filling vacancies, for instance, Unisys' experience has been different. In order to perform a proper evaluation of the Cost/Price factor, NASA was not required to adjust Unisys' probable cost to provide Unisys with a vacancy credit. Vacation The solicitation requires JCAC to provide two, three, or four weeks of paid vacation to employees covered by the Service Contract Act, depending upon length of service. JCAC estimates that it will hire ****** percent of Unisys' current employees and states that it will provide employees with *** ***** of paid vacation. Unisys estimates that between thirty-three and forty percent of its Service Contract Act employees are entitled to receive more than *** ***** of paid vacation. Unisys reasons that JCAC's proposal underestimates the costs it will incur for paid vacation time. In its proposal, JCAC states that it will comply with the terms of the Department of Labor wage determination and the Service Contract Act concerning the amount of vacation time to which employees covered by the Act are entitled. Although JCAC estimates that, if it is awarded the contract, it will hire ****** percent of Unisys' current employees, there is no evidence to establish either which employees will be hired or, of those who will be hired, how much vacation time they are entitled to receive pursuant to the statute, assuming they are Service Contract Act employees. The foundation for Unisys' argument is its estimate that between thirty-three and forty percent of its Service Contract Act employees receive more than two weeks of paid vacation. This estimate is supported only by the testimony of a witness hired by Unisys to provide expert testimony at the hearing and not by the testimony of a witness with knowledge of the facts. Unisys has not demonstrated that JCAC understated its paid vacation costs. Consequently, Unisys has not established that NASA violated statute, regulation, or the terms of the solicitation by making no adjustment to JCAC's proposed vacation costs. Cost of Three Employees The SPC employee testified credibly that JCAC's cost proposal includes the cost of performing the functions of the three positions that JCAC eliminated in its best and final offer. The TEP found nothing wrong with JCAC's proposed staffing levels related to these three positions, and Unisys does not challenge the TEP's findings. Unisys has not established that NASA violated statute, regulation, or the terms of the solicitation by not adjusting JCAC's probable cost to include the costs of the three eliminated positions. Full-Time Employees Although Unisys contends that JCAC and Unisys will need the same number of employees to perform the work required by the contract, there is no support for this contention in the record. Instead, the record establishes that this contract is a mission- type contract and NASA did not dictate to offerors the number of employees required to accomplish the contract work. NASA expected that the staffing levels for offerors would be roughly the same, although offerors would propose different employee skill mixes and different labor categories. In fact, Unisys and JCAC proposed different skill mixes and labor categories. The TEP reviewed JCAC's proposed staffing and made the adjustments that it believed were appropriate, and Unisys does not challenge the TEP's adjustments. Unisys has not demonstrated that JCAC and Unisys would need the same number of full-time employees to perform the work required by the contract. For this reason, Unisys has not established that NASA violated statute, regulation, or the terms of the solicitation when it did not adjust JCAC's probable cost by adding the costs of additional employees. Key Personnel JCAC's cost proposal clearly explains that JCAC anticipated that *** ** *** *** ********* ***** ***** ****** *** ****** ***** ** *** ********. The proposal states that JCAC would try to ******* ***** *********** **** *****-********* ********* *** explains that JCAC's estimated costs are based upon wage surveys. Unisys does not dispute that employee turnover occurs, that JCAC would try to hire qualified replacements, or that JCAC's estimated costs are based upon wage surveys. Although Unisys argues that the TEP evaluated JCAC's proposal as if *** **** ** ****** *** ********* ***** ****** *** *** **** ** *** ********, Unisys does not point to any evidence to establish the TEP's assumptions regarding employee turnover. In addition, Unisys does not challenge the TEP's conclusions. Unisys has not established that NASA violated statute, regulation, or the terms of the solicitation when it did not adjust JCAC's probable cost by adding the costs that JCAC will incur if ***** ** ** ******** ** *** *********. Depreciation - Overhead Ceiling Unisys argues that NASA should not have adjusted Unisys' proposed ceiling overhead rate when it evaluated Unisys' probable cost. Unisys asserts that it would have been bound by the ceiling overhead rate that it proposed and so NASA should have used its proposed ceiling overhead rate in its calculations. The BEP adjusted Unisys' proposed ceiling rate because it adjusted Unisys' bid overhead rate to take depreciation into account and because Unisys' proposal maintained a spread between its bid rate and its ceiling rate. The point of NASA's evaluation was not to select an awardee. Instead, NASA was selecting an offeror for negotiations leading to award. The BEP chair credibly testified that the ceiling rate would have been one of the subjects of negotiations. The BEP properly adjusted Unisys' proposed ceiling rate when the effectiveness of the rate was called into question as a result of the adjustment to Unisys' bid rate. The BEP's adjustment was in keeping with its obligation to consider the cost realism of each offeror's proposal. NASA's actions did not violate statute, regulation, or the terms of the solicitation. Equipment The proposal states that Unisys will provide "up to" ******** of equipment per year. In response to a question from NASA during discussions, Unisys explained that if it did not provide ******** of equipment in one year, the deficit would not be carried forward into succeeding years. In other words, Unisys was not making a hard and fast commitment to purchase ******** of equipment each year. In addition, Unisys was not proposing to purchase equipment for NASA. The equipment would have been carried as an asset on Unisys' balance sheet. Unisys did not establish, by a preponderance of the evidence, that its proposal would provide NASA with a benefit of any specific amount related to equipment. NASA did not violate statute, regulation, or the terms of the solicitation by not giving Unisys credit for ******** of equipment and associated labor. Probable Award Fee and Phase-out Costs Unisys makes the same argument concerning the evaluation of probable award fee as it does concerning the evaluation of phase- out costs. Unisys would have preferred NASA to evaluate these items at less than 100% of the amount proposed by the offerors. The BEP chair testified that, in seventeen years of experience, she has never known fee or phase-out costs to be evaluated at anything other than 100%. Unisys' expert witness testified that he is not aware that the government has ever evaluated fee or phase-out costs at less than 100% and he is not aware of any other expert who has testified that fee or phase-out costs should be evaluated at less than 100%. The solicitation does not state that fee and phase-out costs will be evaluated at something other than 100%. Unisys did not demonstrate, by a preponderance of the evidence, that NASA was required to use a figure other than 100% in its evaluation. NASA did not violate statute, regulation, or the terms of the solicitation by evaluating probable award fee and phase-out costs at 100% of the amount proposed. Phase-in Costs and Phase-out Costs of Unisys' Contract Phase-in costs will not be incurred by Unisys, because Unisys is the incumbent contractor. The solicitation explains that these costs will be the subject of a separate contract. To consider phase-in costs as part of the cost of the support services contract would provide Unisys with an unfair advantage over other offerors. Similarly, the phase-out costs of Unisys' current contract will be incurred only if Unisys, the incumbent contractor, is not selected for award of the new contract. These phase-out costs will be paid as part of Unisys' current contract and not as part of the support services contract. Considering these costs as part of the cost of an award to someone other than Unisys would not be fair because this would always provide Unisys with an advantage over other offerors. The solicitation does not state that the costs of the phase-in contract or the phase-out costs of the current Unisys contract will be included in evaluating the probable cost of the support services contract. No statute or regulation required NASA to evaluate the cost of the support services contract as if it included the costs of other contracts. Unisys did not demonstrate that NASA was required to include these phase-in and phase-out costs in its evaluation of probable cost and did not explain how including such costs would be consistent with ensuring fair competition among offerors. NASA did not violate statute, regulation, or the terms of the solicitation by excluding phase-in costs and phase- out costs of Unisys' current contract from its evaluation of probable cost. Other JCAC Costs Unisys' post-hearing briefs do not address the allegations contained in its complaint. Unisys amended its complaint to delete the conflict of interest allegations which appear to underlie its expert's opinion that NASA should have increased JCAC's probable cost to account for JCAC's implementation of its conflict of interest plan. As for JCAC's inflationary risk, the inflation rate assumed by JCAC is only ***-****** ** *** ******* higher than the rate assumed by Unisys, and there is no evidence to suggest that either assumption is unrealistic. Given the lack of evidence and argument, we deny this ground of protest. Most Advantageous Offer Unisys has met its burden of proving, by a preponderance of the evidence, that NASA's selection of JCAC for negotiations leading to award was not made in compliance with the terms of the solicitation. The Federal Acquisition Regulation provides that agencies "shall evaluate competitive proposals solely on the factors specified in the solicitation" and that the factors established in the solicitation will be used to make the source selection decision. 48 CFR 15.608(a), 15.612(d) (1994). This solicitation contains four evaluation factors. The Mission Suitability factor and the Cost/Price factor are of "essentially equal" importance and are more important than the Other Considerations factor and the Relevant Experience and Past Performance factor. The solicitation states that NASA would select for negotiations leading to award the offeror whose offer will be "most advantageous" to the government. Unisys received higher ratings than did JCAC in both the Other Considerations factor and the Relevant Experience and Past Performance factor. The record establishes that Unisys' higher ratings in these factors did not influence NASA's decision concerning which offer will be more advantageous to the Government. The SSO believed that JCAC's offer did not present any problems in these factors and so the SSO did not afford any importance to the differences between the offerors' ratings. Even though these two factors are not as important as the Cost/Price and Mission Suitability factors, NASA was required to give the evaluation of these factors some weight when it made its source selection decision. The solicitation does not contemplate that NASA would disregard differences between the offerors' ratings for these factors. The differences between the proposals concerning the Mission Suitability factor should have played an important part in the SSO's decision because the solicitation states that this factor is of essentially the same weight as the Cost/Price factor, and there is no significant difference in the offerors' probable costs. In the Mission Suitability factor, Unisys' score is 139 points higher than JCAC's score and the Unisys proposal contains fifteen major strengths, thirty-one minor strengths, no major weaknesses, and five minor weaknesses. In this same factor, JCAC's proposal contains three major strengths, nineteen minor strengths, one major weakness, and ten minor weaknesses. The SSO's decision that JCAC's offer is more advantageous does not take into account the differences between the offerors' Mission Suitability factor point scores which reflect their strengths and weaknesses. The SSO's decision hinged upon the adjectival ratings which mask the specific, concrete differences between the offers. The SSO considered either offeror to be capable of performing the contract work and, as a result, the Mission Suitability factor played essentially no part in the SSO's decision and the Cost/Price factor assumed an importance to which offerors were not alerted by the solicitation. The record also establishes that the SSO's decision did not result from an analysis, which the parties refer to as a cost/technical tradeoff, of whether Unisys' superior technical proposal is worth any additional cost. Such a tradeoff between the cost and technical elements of a proposal is one means of deciding which offer will be most advantageous to the Government. The TEP determined that some of Unisys' major strengths have the potential to improve efficiency, reduce risk, and minimize costs. The SSO decided that, if these advantages could not be quantified, they could not make up for Unisys' higher cost. The TEP, however, did not determine that Unisys' technical advantages have no value. Rather, the TEP determined that it could not translate those advantages into dollars. When the TEP could not assign a dollar value to Unisys' technical advantages, the SSO did not attempt to determine whether those advantages could account for Unisys' higher cost. Instead, the SSO chose to negotiate with JCAC because JCAC proposed a lower cost. Regardless of whether the TEP could quantify the dollar value of Unisys' technical advantages, the SSO was required to determine whether Unisys' or JCAC's offer will be more advantageous to the Government and this required the SSO to take into account both cost and technical merit. The evidence does not establish why the SSO decided that Unisys' advantages were not worth an additional cost or how the SSO decided that Unisys' offer was less advantageous than JCAC's offer. NASA points out that the SSO listened to the SEB's presentation of the facts concerning each offeror's proposal. The record establishes that the SSO listened to the March 8, 1995 presentation and discussed the SEB's findings. But, the record does not establish how the SSO took the information presented by the SEB into account when he made his decision to conduct negotiations with JCAC. The fact that the SSO listened to the SEB does not provide any meaningful description or reasoned analysis of how the SSO arrived at his decision. Simply listening to the presentation was not enough. The SSO was required to assess independently the information presented by the SEB and to arrive at a reasoned conclusion as to whether JCAC's or Unisys' offer will be more advantageous to the Government. NASA's generalized showing that the SSO met with the SEB does not outweigh the evidence adduced by Unisys. NASA emphasizes that source selection officials are entitled to exercise discretion when they perform their duties. We agree. The extent, however, to which a source selection official may exercise discretion is defined and limited by the terms of the solicitation. This solicitation states that four evaluation factors would be considered. The evidence is that the evaluated differences between JCAC and Unisys in three of those factors played no part in the SSO's decision. The solicitation states that the Mission Suitability factor and the Cost/Price factor were of essentially equal importance. The evidence is that the SSO's decision fails to take into account the offerors' Mission Suitability differences and, as a consequence, the Cost/Price factor became by far the most weighty evaluation factor. Finally, the solicitation states that NASA would negotiate with the offeror whose offer will be most advantageous to the Government. The evidence is that the SSO, instead of weighing technical advantages and cost in order to arrive at a decision concerning the relative advantages of the offers, chose to negotiate with the less costly offeror. Perhaps JCAC's offer will be more advantageous to the Government. Perhaps not. We simply hold that, based upon the record before us, Unisys has established that NASA did not abide by the terms of the solicitation when it chose JCAC for negotiations. Decision The protest is GRANTED. There is no need for NASA to conduct another technical evaluation or another price evaluation, although it may do so if it wishes. NASA is directed to determine, consistent with the terms of the solicitation, whether JCAC's offer or Unisys' offer will be more advantageous to the Government, to articulate and document the basis for its determination as required by 48 CFR 15.612(d), or otherwise to proceed with the procurement consistent with statute and regulation. ______________________________ MARTHA H. DeGRAFF Board Judge We concur: _______________________________ _______________________________ MARY ELLEN COSTER WILLIAMS ALLAN H. GOODMAN Board Judge Board Judge