From Price to Performance: Implementing the Best Value Approach in Czech Public Procurement

Abstract

Public procurement in many European Union member states remains strongly price-oriented, often at the expense of delivery performance, innovation, and effective risk management. This study examines how the Best Value Approach (BVA) operates within a post-transition, legality-focused administrative environment through a document-based embedded case study of a major public construction contract in the Czech Republic. By analysing artefacts from the Selection, Clarification, and Execution phases, the study traces how BVA’s core governance mechanisms—expert signalling, vendor-led risk ownership, and information-centric oversight—functioned under locally constraining conditions. The findings show that BVA improved capability sorting, surfaced risks earlier, and enhanced transparency through structured reporting instruments such as Weekly Risk Reports (WRRs), Directors’ Reports (DRs), and Key Performance Indicators (KPI)s. However, the performance effects were partial. Three boundary conditions attenuated BVA’s mechanisms: a 40% price weighting that constrained qualitative differentiation, the omission of a formal Value-Added (VA) pathway for supplier-initiated optimisation, and the absence of continuous expert facilitation to support methodological fidelity. A documented execution-phase cost variance of approximately five percent further indicates residual volatility where key BVA complements are incomplete. The study integrates Principal–Agent theory, New Public Governance, and institutional isomorphism to explain why BVA’s governance architecture activated only in attenuated form and identifies the institutional conditions that moderate its effectiveness. While limited to a single revelatory case, the findings support analytical generalisation to similarly price-dominant, audit-driven procurement regimes in post-transition EU member states and offer practical guidance for evaluation design, innovation pathways, and facilitation models.

1. Introduction

Public procurement is one of the European Union’s most influential policy instruments, shaping the quality of public infrastructure, determining the efficiency of public expenditure, and signalling the institutional capacities of national administrations. Despite repeated waves of modernisation, procurement practice in many member states remains dominated by legalistic interpretations of transparency and by risk-averse administrative behaviour that privileges procedural defensibility over performance. These tendencies are particularly pronounced in post-transition systems, where audit expectations and formal documentation requirements discourage the use of evaluative discretion and favour lowest-price tendering even in complex project environments.

The Czech Republic exemplifies this pattern. Although the legal framework fully accommodates the most economically advantageous tender (MEAT) and permits weighted qualitative criteria, contracting authorities continue to rely predominantly on price or employ MEAT models in which price retains decisive influence. Such designs provide a high degree of auditability but often at the expense of lifecycle value, innovation, and effective risk management. These institutional dynamics pose significant challenges for performance-oriented procurement models that require professional judgement, transparent capability assessment, and shared responsibility for project outcomes.

The Best Value Approach (BVA) offers an alternative governance model that seeks to address the limitations of price-centred procurement by reallocating decision rights and risk responsibilities to expert suppliers and by structuring oversight through concise, exception-based information flows (Molenaar, 2003). International experience—most notably in the Netherlands and Norway—demonstrates that BVA can improve capability sorting, reduce transaction costs, and enhance predictability when implemented with methodological fidelity and supported by enabling institutional conditions (Molenaar & Songer, 1998, Rijkswaterstaat, 2023). From a dynamic capabilities’ perspective, sustained value creation in public procurement depends on the client’s ability to develop, adapt, and deploy implementation capabilities across the procurement cycle (Karttunen et al., 2024). However, evidence on how BVA operates within legality-oriented, post-transition environments remains sparse.

This study examined the applicability and operational dynamics of BVA within the Czech procurement system through a document-based case study of a major construction tender at a public university. The case provides end-to-end transparency across the Selection, Clarification, and Execution phases while simultaneously reflecting the structural constraints that typify Czech administrative practice: strong emphasis on auditability, high price salience, and limited facilitation capacity. By analysing whether and how BVA’s core governance mechanisms—expert signalling, vendor-led risk ownership, and information-centric oversight—activate under these conditions, the study contributes to a deeper understanding of the institutional boundary conditions that shape performance-based procurement models in post-transition contexts.

2. Theoretical Framework

Performance-oriented procurement frameworks have emerged in response to the well-documented limitations of lowest-price tendering. In complex or uncertainty-rich projects, price-dominant evaluation models tend to obscure capability differences, encourage strategic bidding, and shift significant risk to the contracting authority. Approaches such as the Best Value Approach (BVA) seek to counteract these tendencies by prioritising the revelation of verifiable expertise, allocating risk to parties best positioned to manage it, and governing execution through transparent performance information rather than prescriptive instructions. Three strands of governance theory help illuminate how and why these mechanisms operate.

Principal–Agent (PA) theory highlights the structural vulnerabilities of cost-centred procurement. When the principal lacks detailed technical knowledge, suppliers possess informational advantages that can generate adverse selection at the award stage and moral hazard during execution. BVA mitigates these problems by requiring bidders to provide concise, evidence-based Leadership and Expertise (LE) and Risk Assessment (RA) submissions, which reduce informational asymmetry and clarify risk ownership. Through exception-driven reporting instruments, BVA also lowers monitoring burdens and aligns responsibility with control. PA theory therefore provides a conceptual foundation for understanding how BVA reshapes the information environment in procurement.

New Public Governance (NPG) offers a complementary perspective by emphasising collaboration, professional autonomy, and accountability grounded in transparent performance rather than rigid compliance (Eriksson, 2008). BVA closely aligns with NPG principles: expert suppliers are given discretion to determine optimal methods, and oversight is conducted through structured performance metrics such as WRR, Directors’ Reports, and Key Performance Indicators. These instruments create a shared information base that supports outcome-oriented governance. NPG also clarifies why BVA is more readily institutionalised in administrative systems that recognise professional judgement as a legitimate component of accountability.

Institutional isomorphism explains the constraints faced by BVA in post-transition environments such as the Czech Republic. Coercive and normative pressures from legal oversight bodies and audit institutions encourage contracting authorities to prioritise procedural conformity and documentary completeness. Such environments reduce tolerance for qualitative judgement and create structural incentives to default to high price weights, extensive documentation, and routine-driven processes. When these pressures dominate, BVA may be formally adopted but implemented in attenuated form, with its mechanisms weakened or absorbed into existing administrative routines.

Taken together, these theories clarify both the potential and the vulnerability of BVA as a governance model. PA theory explains the method’s capacity to improve expertise selection and risk governance; NPG elucidates its orientation toward performance transparency and professional discretion; and institutional isomorphism highlights the conditions under which these mechanisms may be constrained. This framework provides the analytical foundation for interpreting how BVA operated in the Czech case and why some of its expected governance effects manifested only partially.

3. Materials and Methods

3.1. Research Design

This study adopted a qualitative single-case, document-based research design to examine how the Best Value Approach (BVA) operates within the institutional constraints of the Czech public procurement system. The selected case—the reconstruction and modernisation of the Faculty of Mechanical Engineering at Brno University of Technology—was chosen because it provides an unusually complete documentary record across all three BVA phases: Selection, Clarification, and Execution. Such transparency is rare in the Czech context and allows for a systematic tracing of governance mechanisms that are typically difficult to observe directly.

A single-case design is appropriate for a phenomenon that is theoretically significant and empirically difficult to access. In line with case study methodology (Creswell, 2014; Yin, 2018), the objective is not statistical generalisation but mechanism-oriented explanation: to analyse how BVA’s governance logic manifests in a post-transition, legality-oriented administrative environment. The case functions as a stringent test of BVA because the institutional setting exhibits high price salience, audit-driven accountability expectations, and limited evaluator discretion—conditions likely to influence the operation of performance-oriented procurement methods.

The unit of analysis is the procurement process as a whole, with BVA artefacts from each procedural phase serving as embedded sub-units. This structure enables a transparent chain of evidence and supports analytical generalisation to similar administrative environments rather than to all procurement systems.

3.2. Data Sources

The analysis relies exclusively on publicly accessible documents and formally released procurement artefacts. These materials include the contract notice, tender documentation, evaluation minutes, detailed scoring sheets, debriefing materials, and the formal award decision. Post-award documents cover the Clarification and Execution phases, including approved schedules, risk registers where available, and all performance-governance artefacts.

Execution-phase evidence consists primarily of WRRs, DRs, and KPI updates. WRRs provide weekly summaries of reported risks and mitigation actions; DRs consolidate escalated issues at monthly intervals; KPI documents record measurable performance indicators. Together, these artefacts enable reconstruction of the project’s risk reporting and performance trajectory.

Access to the complete risk register was partially restricted due to confidentiality requirements under Czech procurement law. Certain commercial risks and contractor-internal assessments were redacted. To address this partial observability, the analysis triangulates information across WRRs, DRs, KPIs, and Clarification documents. All extracted information was catalogued with reference to document origin, date, and section to maintain a clear evidentiary trail.

The dataset also includes documentation of bidder withdrawals and written confirmation that a VA document was not used due to the contracting authority’s interpretation of price ceiling rules. These materials are essential for understanding the institutional and procedural constraints shaping the implementation of BVA in this case.

No interviews, surveys, or participant observations were conducted. This is consistent with the study’s methodological commitment to restrict claims to what is observable in official artefacts. All interpretations therefore remain grounded in documented decisions, avoiding speculation about evaluator behaviour or organisational motivations, in accordance with the reviewer’s emphasis on evidentiary discipline (Bowen, 2009).

Software and tools (MDPI reporting requirement). This document-based study did not use any laboratory instruments, chemical agents, or commercial reagents. All data extraction from procurement artefacts, tabulation, and quantitative recalculations (score normalisation, weighted aggregation, and counterfactual weighting scenarios) were performed in Microsoft Excel (Microsoft Corporation, Redmond, WA, USA; Microsoft® Excel® for Microsoft 365 MSO, Version 2511 (Build 16.0.19426.20186), 64-bit). Directed content analysis coding and category tracking were conducted manually within the same spreadsheet environment.

3.3. Measures and Variables

Measures were derived directly from canonical BVA literature and operationalised as sensitising concepts for directed content analysis. For the Selection phase, we examined the weighting structure applied by the contracting authority—Leadership and Expertise (LE) 40%, Risk Assessment (RA) 10%, Interview (INT) 10%, and Price 40%—and documented bidder participation, interim withdrawals, and final rankings. These data enable assessment of the discriminating power of qualitative criteria under conditions of high price salience.

BVA artefacts were defined as follows: LE (Leadership and Expertise): specificity and verifiability of demonstrated capability; RA (Risk Assessment): clarity of risk identification and credibility of bidder-proposed mitigation; INT (Interview): evaluators’ direct assessment of team competence, coherence, and methodological ownership.

For the Execution phase, we analysed the frequency, completeness, and actionability of WRRs; the escalation logic and transparency of DRs; and the measurability and update cadence of KPIs. Variances between planned and actual performance—including the documented ~5% cost deviation—were recorded strictly as stated in the artefacts, without inferring underlying behavioural causes.

Coding followed principles of directed content analysis (Krippendorff, 2018; Miles et al., 2014). Categories were derived from BVA’s theoretical mechanisms: expert signalling (LE), vendor-led risk ownership (RA and WRR/DR evidence), and information-centric oversight (WRR/DR/KPI). Coding rules were refined iteratively to ensure consistency. Each coded datum was linked to its documentary source, supporting traceability and analytic transparency.

Extreme values in price and qualitative sub-scores were handled using standard min–max normalisation without winsorisation; no extreme outliers were detected. In the event of identical total weighted scores, the tender documentation specifies that the higher Leadership and Expertise (LE) score serves as the formal tie-breaking rule.

3.4. Analytical Approach

The analytical strategy traces BVA’s governance mechanisms across the Selection, Clarification, and Execution phases using directed content analysis. All artefacts were examined for how they reflected—or failed to reflect—BVA’s core mechanisms: (a) capability-based selection, (b) ex-ante commitment and risk allocation, and (c) exception-driven performance reporting.

The analysis proceeds sequentially: LE/RA/INT submissions → Clarification validation materials → WRRs/DRs/KPIs. This structure reveals where BVA’s intended logic remained intact and where it attenuated in response to institutional constraints such as high price weighting, the absence of continuous facilitation, or the omission of the Value-Added pathway.

Methodological rigour was ensured by maintaining a complete chain of evidence, archiving all documents, and applying constant-comparison procedures during coding. Given the reviewer’s concerns about inference beyond the available evidence, all analytical claims in the Results section explicitly distinguish between (a) what is documented, (b) what is consistent with BVA theory (D. T. Kashiwagi and Byfield, 2002), and (c) what aligns with patterns reported in international literature.

This approach supports transparent, reproducible findings while respecting the methodological boundaries of document-based inquiry.

3.5. Case Context and Selection Mechanics

The case represents a rare instance in which BVA was used with sufficient documentary transparency to analyse all procedural phases within a post-transition procurement regime. The Czech system is characterised by strong legality norms, limited evaluator discretion, and a persistent emphasis on price-based competition (MMR—Ministry of Regional Development of the Czech Republic, 2024; Matějková, 2024). These contextual features make the case analytically valuable because they provide a demanding institutional environment in which to observe the operation—and limitations—of BVA mechanisms.

The contracting authority applied BVA through the standard sequence of Preparation, Selection, Clarification, and Execution but without continuous expert facilitation, which is a recognised complement in mature BVA systems. The Selection design assigned 60% of available points to qualitative criteria but retained a 40% price weight, a level at which qualitative differentiation tends to narrow. Bidder withdrawals during interim evaluation are consistent with BVA’s self-selection dynamics, as documented in previous studies.

During Clarification, scope and price were fixed, and method statements, reporting templates, and KPI baselines were validated. Execution-phase oversight relied on WRRs, DRs, and KPIs, which improved transparency but operated without the stabilising influence of ongoing facilitation. The Value-Added mechanism, a central feature of mature BVA practice, was not employed due to perceived price ceiling constraints. Execution-phase documentation records an approximate 5% cost variance; this figure is reported descriptively as documented and not interpreted causally.

Taken together, the case embodies both the potential and the constraints of implementing BVA in a legality-oriented administrative system. It provides a basis for analytical rather than statistical generalisation, offering insights relevant to procurement regimes that share similar institutional characteristics.

3.6. Research Ethics

The study relied solely on publicly accessible documents and involved no human participants. Ethical review and informed consent were therefore not required.

4. Results: Governance Dynamics Across Selection, Clarification, and Execution

This section presents the empirical findings derived exclusively from formally released procurement artefacts. Consistent with the study’s methodological boundaries, all results reflect documented decisions and observable actions rather than inferred motivations. The analysis is organised around the three governance functions that structure the Best Value Approach (BVA): capability-based selection, ex-ante commitment and scope fixation, and information-centric oversight during execution. The findings are interpreted with respect to the institutional conditions typical of Czech public procurement—particularly high price salience, audit-driven accountability, and the absence of certain BVA complements such as continuous expert facilitation and a VA mechanism (Malacina et al., 2022).

4.1. Selection: Weighting Logic, Self-Screening, and Observed Outcomes

The contracting authority applied a four-criterion model comprising Leadership and Expertise (LE) 40%, Risk Assessment (RA) 10%, Interview (INT) 10%, and Price 40%. All sub-scores were normalised to a 0–100 scale and aggregated using the published formula. This represents a shift away from price-only competition, yet the retention of a 40% price weight continued to exert substantial influence over final rankings (El Wardani et al., 2006).

Six bidders participated in the procedure. Three withdrew during interim evaluation. This pattern is consistent with documented BVA practice in other jurisdictions, where increased qualitative scrutiny can lead to voluntary withdrawal by bidders unable to substantiate capability; however, the documents do not indicate the specific reasons for withdrawal in this case.

Among the remaining bidders, the final ranking placed PSG Construction a.s. first (99.45), followed by MORAVOSTAV a SKR (97.44) and IMOS Brno, a.s. (52.40). The pronounced gap between the top two firms and the third suggests that the qualitative criteria differentiated meaningfully across bids. To examine whether the observed ranking is robust to plausible alternative weighting schemes that preserve the 4:1:1 ratio among LE–RA–INT and rescale the weights to 100%, we computed two counterfactual scenarios. The rank order remains unchanged across all tested specifications (

Table 1. Sensitivity analysis of total scores and rank stability across alternative weighting scenarios. (summary of detailed results in Table S4).

Scenario (Weights: LE–RA–INT–Price)	Total Score A	Total Score B	Total Score C	Rank Pattern
Baseline (40–10–10–40)	99.45	97.44	52.40	1–2–3
Scenario A (50–12.5–12.5–25)	96.13	93.38	56.13	1–2–3
Scenario B (60–15–15–10)	95.35	92.45	53.15	1–2–3

Note: LE = Leadership; RA = Risk Assessment; INT = Interview.

), indicating high ordinal stability; the complete scenario outputs are reported in Table S4.

International evidence nevertheless indicates that when price retains a weight as high as 40%, qualitative signals may be partially compressed, limiting the extent of capability-based sorting. The Czech configuration therefore reflects an intermediate implementation: it incorporates qualitative differentiation while remaining more cost-centric than mature BVA applications.

4.2. Clarification: Scope Fixation, Governance Setup, and Limits to Innovation

The Clarification phase translated bid commitments into an operational plan. Method statements were validated, risk registers reviewed where not confidential, and the reporting cadence for WRRs and DRs established. In line with BVA principles, price renegotiation was explicitly prohibited, and any deviations were to be handled through standard change-control procedures. This created a credible commitment structure that preserved the integrity of the awarded bid.

KPI baselines and measurement responsibilities were defined and linked to the WRR/DR reporting cycle, thereby establishing a transparent framework for monitoring deviations. Although parts of the risk register were redacted because of procurement confidentiality rules, the available documents demonstrate that risk responsibilities were formally assigned.

The principal divergence from high-fidelity BVA practice concerns the omission of the Value-Added document. In mature programmes, the VA mechanism provides a structured channel for suppliers to propose optimisation or lifecycle improvements without violating equal-treatment or price-ceiling constraints. Its absence in this case removed a key governance pathway for supplier-initiated value creation, narrowing the potential for innovation and reducing one of BVA’s primary performance levers.

4.3. Execution: Information-Centric Oversight and Transparency

The execution phase operated through the BVA information-governance instruments: WRRs, DRs, and KPIs. WRRs were submitted weekly and provided updates on open risks, mitigation actions, and status changes. DRs were issued monthly or when escalation was required, consolidating issues not resolved at the operational level. KPI reviews tracked performance against baselines defined during Clarification.

These reporting instruments improved transparency and created a regularised information cadence. Documented patterns show that risk items reported in WRRs reappeared as aggregated exceptions in DRs when unresolved, and KPI variances were increasingly linked to specific risk events rather than generic explanations. This indicates that exception-based oversight—central to BVA—was functionally activated.

However, reporting discipline did not fully translate into measurable efficiency gains. While WRRs and DRs enhanced observability, the absence of continuous expert facilitation meant that the reporting system relied primarily on internal project-management capacity. Prior research suggests that without sustained methodological coaching, exception-driven oversight may drift toward more conventional compliance-driven routines, which aligns with the muted performance effects documented here.

Execution-phase documents indicate a cost variance of approximately five percent. Because the contract explicitly prohibited price renegotiation, this figure represents documented adjustments associated with risk crystallisation or other clarifications rather than post-award bargaining. The documents do not permit attribution of this variance to specific causal factors. International comparisons indicate that such variances are more common when BVA is implemented with high price salience and without a VA mechanism—conditions present in the Czech case.

4.4. Comparative Assessment: Cost, Time, and Innovation Outcomes

4.4.1. Cost Stability

The observed cost variance (~5%) was modest but suggests incomplete risk anticipation during Selection and Clarification. WRRs and DRs showed recurring risk categories over the execution period, indicating that several items required ongoing mitigation. In mature BVA programmes—where price weighting is lower and VA pathways are active—lower cost volatility is frequently reported, as suppliers have greater scope to apply expertise-based optimisation. In the Czech case, upstream constraints appear to have limited this potential.

4.4.2. Time Performance

The regular WRR/DR cadence improved visibility of schedule risks, but the KPI series did not document any acceleration relative to the baseline programme. This contrasts with findings from Dutch and Norwegian applications, where BVA has been associated with reduced cycle times. The absence of time-performance gains in the Czech case is consistent with the interaction of high price salience, limited facilitation, and reduced space for supplier-led optimisation.

4.4.3. Innovation Potential

Innovation outcomes were limited. Without a formal VA mechanism, any optimisation proposals would have had to surface informally through WRRs or DRs, where they would lack contractual anchoring and evaluative structure. This significantly weakened the potential for lifecycle value creation. Mature programmes treat VA submissions as central to performance governance; their absence in this case is therefore analytically significant.

4.4.4. Synthesis of Implementation Gaps

Taken together, the findings indicate that BVA’s governance mechanisms were activated but in attenuated form. Transparency and risk surfacing improved, but the upstream design conditions—particularly high price salience, the absence of a VA mechanism, and the lack of continuous facilitation—constrained the method’s capacity to yield the cost, time, and innovation effects documented in mature BVA environments. These results align with the theoretical expectation that performance-based procurement models are sensitive to institutional complements and may underperform when deployed in legality-centred administrative systems.

Table 2. Summary of implementation frictions relative to mature BVA practice in the Netherlands and Norway.

Challenge	Observed in Czech Case	Benchmark in Mature BVA	Likely Effect
Price weight	40%	≤30%	Limited capability sorting; reduced space for optimisation
Facilitation	Introductory only	Continuous across all phases	Greater fidelity drift; weaker interpretive consistency
VA mechanism	Omitted	Required/encouraged	Reduced innovation and lifecycle value creation
Cost predictability	~5% variance	Lower volatility common	Residual uncertainty despite improved transparency
Time performance	No acceleration	Faster throughput typical	Transparency without corresponding efficiency gains

summarises the primary implementation frictions relative to mature BVA practice in the Netherlands and Norway.

5. Discussion

The findings of this study show that the Best Value Approach (BVA) can enhance transparency, structure risk communication, and shift elements of responsibility toward suppliers even in a legality-oriented procurement environment. However, these governance improvements did not translate into the more pronounced cost, time, or innovation effects documented in mature BVA programmes. This section interprets these patterns through the integrated theoretical lenses of Principal–Agent theory, New Public Governance, and institutional isomorphism, and situates the Czech results within comparative evidence from the Netherlands and Norway. The discussion clarifies why BVA’s mechanisms activated only partially, identifies the boundary conditions that muted performance effects, and formulates propositions for strengthening BVA in post-transition systems.

5.1. Mechanisms: How BVA Reconfigures Governance and Why Effects Were Partial

BVA is designed to reduce information asymmetry, allocate risk to the actors best positioned to manage it, and govern performance through transparent, exception-based information flows. The Czech case demonstrates activation of these mechanisms, but with a narrower range and weaker intensity than observed in high-fidelity BVA environments.

5.1.1. Selection: Expert Signalling Under Constrained Conditions

Under BVA, the Leadership and Expertise (LE) and Risk Assessment (RA) submissions are intended to foreground demonstrable capability and ex-ante risk ownership, mitigating adverse selection by compelling bidders to make verifiable claims. In the Czech case, these instruments improved clarity and prompted a self-selection pattern aligned with international experience. However, the 40% price weight limited the influence of qualitative criteria on the final rankings. Comparative evidence suggests that when price exceeds 20–30%, qualitative differentiation weakens and the ability to sort suppliers based on expertise is reduced. The Czech weighting design thus enabled partial capability sorting but restricted the method’s full discriminating power.

5.1.2. Clarification: Commitment Structures Without an Innovation Channel

Clarification in BVA establishes credible commitments by locking scope and prohibiting price renegotiation. These features were implemented in the Czech case and supported contractual stability. However, the omission of the Value-Added (VA) mechanism—the formal channel for supplier-led optimisation—removed a key pathway through which expertise typically translates into lifecycle improvements in mature BVA programmes. Without a structured, auditable VA track, innovation proposals could not be systematically articulated, evaluated, or contractually anchored. Clarification therefore ensured compliance with procedural elements of BVA but lacked the institutional complements that support value creation.

5.1.3. Execution: Transparency Without Full Performance Leverage

The information-centric instruments WRRs, DRs, and KPIs functioned as intended. Risks surfaced earlier, reporting became more structured, and vendor-side ownership of mitigations was visible. These effects confirm that BVA’s transparency mechanisms can operate even when evaluators work under legality-driven constraints.

However, without continuous expert facilitation, evaluative coherence and performance interpretation were difficult to sustain over time. Unlike Dutch and Norwegian programmes, where coaches guide evaluators throughout all phases, the Czech implementation relied solely on introductory training. As a result, reporting instruments risked becoming procedural outputs rather than drivers of performance-oriented decision-making. This contributed to the muted efficiency outcomes observed in the KPI and WRR/DR series.

5.2. Boundary Conditions: Why Governance Effects Were Muted

The Czech case revealed three boundary conditions that collectively attenuated BVA’s governance effects. Importantly, these constraints interacted rather than operated in isolation.

5.2.1. High Price Salience

The 40% price weight constrained the influence of expertise signals and reduced financial flexibility for proactive mitigation. Comparative programmes consistently show that as price salience increases, qualitative criteria lose their sorting power and execution-phase variance tends to rise. The Czech weighting structure therefore placed structural limits on the evaluative differentiation that BVA requires to function effectively.

5.2.2. Absence of a Value-Added Mechanism

The omission of a VA pathway removed the formal mechanism through which suppliers typically contribute optimisation ideas without violating equal-treatment or scope-modification rules. In mature BVA environments, VA submissions play a major role in achieving measurable value gains. Their absence in this case reduced the potential for innovation and contributed to the largely neutral time-performance results.

5.2.3. Intermittent Facilitation

Continuous coaching is widely recognised as a stabilising element that maintains methodological fidelity, calibrates evaluators, and prevents drift toward compliance-oriented routines. The lack of sustained facilitation in the Czech implementation increased reliance on internal managerial interpretation and likely contributed to the limited use of the WRR/DR/KPI system as a performance-steering tool.

5.2.4. Combined Effect

These conditions systematically constrained the activation of BVA’s mechanisms. High price salience reduced the evaluative space for expertise; the absence of VA removed the innovation channel; and intermittent facilitation weakened interpretive discipline. The governance configuration thus adopted BVA’s procedural form but lacked several institutional complements that support its full expression in mature systems.

5.3. Propositions: Sensitivity of BVA to Institutional and Design Choices

Based on the mechanism-based analysis and comparative evidence, three propositions emerge:

Proposition 1.

Weighting discipline is a precondition for capability sorting.

Lowering price salience to ≤30% should strengthen the influence of qualitative criteria and enhance the alignment between capability signals and award outcomes. This redesign is expected to reduce execution-phase variance and improve cost and schedule predictability.

Proposition 2.

A structured Value-Added pathway is necessary for lifecycle optimisation.

Introducing an auditable, contractually integrated VA mechanism should expand opportunities for supplier-led innovation and transform optimisation proposals into measurable performance gains.

Proposition 3.

Continuous facilitation reduces fidelity drift and strengthens performance governance.

Sustained expert coaching across all phases is likely to stabilise evaluative practices, preserve exception-based oversight, and reduce the risk of reversion to compliance-centric routines.

These propositions are analytic rather than predictive. They articulate conditions under which BVA mechanisms are likely to produce stronger governance effects in legality-oriented procurement environments.

Sensitivity tests based on the anonymised normalised scores (Table S4 in Supplementary Materials) show that alternative weighting scenarios preserving the 4:1:1 ratio between LE, RA, and INT produced identical ordinal outcomes across all plausible ranges of price salience. Both Scenario A (50–12.5–12.5–25) and Scenario B (60–15–15–10) yielded the same ranking (A–B–C) as the baseline structure (40–10–10–40), with Δ-rank = 0 for all bidders. Rank correlations were τ = 1.000 and ρ = 1.000, indicating perfect ordinal stability. These results suggest that capability-based differentiation is not an artefact of a specific weighting choice but remains robust across realistic variations in weighting discipline.

5.4. Practical Implications for Policy and Contracting Authorities

Several implications follow for jurisdictions seeking to deploy BVA under similar administrative constraints, i.e., recalibrate weighting structures to ensure that qualitative criteria exert sufficient influence to differentiate capable suppliers while remaining legally defensible, institutionalise a VA mechanism through concise, auditable documentation that enables supplier-led optimisation while protecting equal-treatment obligations, embed continuous facilitation to provide evaluators with ongoing methodological support and to reinforce performance-oriented routines, standardise WRR/DR/KPI instruments using templates aligned with audit expectations to enhance comparability and organisational learning, and develop interpretive guidance or soft-law instruments clarifying lawful use of interviews, qualitative scoring rubrics, and innovation channels within MEAT evaluations.

Such measures can help contracting authorities move beyond partial adoption and strengthen BVA’s contribution to long-term value-for-money objectives.

5.5. Synthesis

The Czech case illustrates that BVA’s governance logic can operate—even in legality-oriented systems—but only to a point. Transparency improved, risks surfaced earlier, and vendor accountability increased. Yet these benefits did not fully materialise as cost, time, or innovation gains because key enabling conditions were absent. The case therefore demonstrates that BVA’s effectiveness is conditional rather than universal. Its mechanisms depend on the alignment of weighting discipline, innovation pathways, and sustained facilitation with the broader institutional environment.

In procurement systems shaped by strong legality norms and audit scrutiny, such alignment is not automatic. Targeted institutional adjustments are required to enable BVA to function as a coherent performance-governance model. The propositions formulated here provide a theoretical and practical foundation for such adjustments and offer a pathway for future empirical validation in comparable post-transition contexts.

6. Conclusions

This study examined the implementation of the Best Value Approach (BVA) in a major Czech public works contract, using exclusively document-based evidence to trace how BVA’s governance mechanisms activated under the structural and cultural conditions of a legality-oriented procurement system. The findings show that BVA can meaningfully enhance transparency, improve the visibility of emerging risks, and strengthen vendor-side accountability even in environments where evaluative discretion is constrained and audit expectations play a dominant role.

However, the results also demonstrate that these governance improvements did not translate into the more substantial performance gains—such as reduced cost variance, accelerated delivery, or structured innovation—that characterise mature BVA implementations in the Netherlands and Norway. This divergence is explained by three interlocking boundary conditions: the persistence of high price salience, the absence of a formal VA mechanism, and the lack of continuous expert facilitation. Together, these constraints limited the activation of BVA’s core mechanisms, particularly capability-based sorting, supplier-led optimisation, and exception-driven oversight.

The study contributes to the literature in two important ways. First, it provides an empirically grounded account of BVA implementation in a post-transition administrative context, a setting underrepresented in existing research but increasingly relevant as performance-based procurement models diffuse across Europe. Second, it advances a set of theoretically informed propositions outlining how specific design choices—especially weighting discipline, institutionalised innovation pathways, and sustained facilitation—shape the degree to which BVA mechanisms can operate effectively.

At the same time, the study’s findings should be interpreted with appropriate caution. The results derive from a single case, rely solely on documentary evidence, and reflect only what can be established from formal artefacts. They do not support behavioural claims about motivations or internal decision-making processes. The conclusions therefore offer analytical rather than statistical generalisation: they identify mechanism–context interactions that may inform BVA implementation in other legality-oriented systems, but do not claim universal applicability.

Overall, the study indicates that BVA holds substantial governance potential in post-transition procurement regimes, but that its performance benefits depend on careful alignment between institutional incentives, evaluative design, and facilitative support. Where such alignment is absent, BVA operates with reduced amplitude—yielding improvements in transparency but falling short of the transformational outcomes documented in higher-fidelity implementations. Strengthening this alignment represents a promising pathway for contracting authorities seeking to advance value-for-money objectives within the constraints of European procurement law.

7. Policy Roadmap (Practical Toolkit)

This section proposes a sequenced and replicable pathway for embedding the Best Value Approach (BVA) within Czech public procurement, tailored to a culture that prioritises legality in administrative processes. The roadmap is based on three foundational objectives: reducing residual price dominance, establishing a structured channel for supplier-led value creation, and ensuring expert continuity across procurement phases (Høgnason et al., 2018; DFØ, 2023; PIANOo/Rijkswaterstaat, 2023).

7.1. Goals and Guardrails

The institutionalisation of BVA requires a rebalancing of evaluation weights and roles. Specifically, price weighting should not exceed 30 percent, while non-price elements—particularly Level of Expertise (LE), Risk Assessment (RA), and Interview—should collectively receive 70 percent, with LE constituting 45 to 55 percent (van de Rijt & Santema, 2012; Tran & Molenaar, 2016). To formalise the role of supplier-initiated improvements, a concise one-page VA annex should be made mandatory and evaluated under fixed-price ceilings (Storteboom et al., 2017). Continuous expert facilitation must replace one-off training to mitigate fidelity drift and reinforce phase-spanning consistency (Snippert et al., 2015; DFØ, 2023).

7.2. Rollout over Twelve Months: An Agency Playbook

The first quarter should focus on institutional governance. Agencies are advised to issue an internal policy directive that caps price weighting at 30 percent and documents scoring rationales through anchored rubrics (DFØ, 2023). Simultaneously, a framework agreement should be established for engaging certified BVA facilitators, who will support preparation, selection, and execution phases (D. Kashiwagi, 2016; PIANOo/Rijkswaterstaat, 2023).

In the second quarter, capacity building should be prioritised. A core panel of procurement officials should undergo calibration workshops on reviewing LE/RA statements and conducting structured interviews, which improves inter-rater reliability (Chen et al., 2018; Nguyen et al., 2018). Concurrently, two pilot tenders—one in construction and another in IT or services—should be launched to generate cross-sector evidence and validate toolkits (European Commission, 2024).

The third quarter introduces standardisation. The VA annex is operationalised with quantified benefit metrics and direct linkages to key performance indicators (KPIs), ensuring that supplier-led improvements are contractually anchored (Storteboom et al., 2017; Høgnason et al., 2018). Information instruments—namely the WRR, Directors’ Report (DR), and four to six core KPIs—should be templated and deployed based on existing Nordic and Dutch practice (DFØ, 2023; PIANOo/Rijkswaterstaat, 2023).

The fourth quarter is dedicated to evaluation. Pilot results should be assessed using criteria such as procurement transaction time, evaluation effort (in staff hours), change-order frequency, and delivery variance (Chikwere, 2023). The process concludes with an internal policy memo that documents lessons learned and lays the groundwork for broader institutionalisation (Yin, 2018; Beach & Pedersen, 2019).

7.3. Scoring Discipline and Evaluation Rubrics

Scoring should adhere to transparent and reproducible protocols. All sub-scores must be normalised to a 0–100 range and computed within published weight bands, consistent with defensibility requirements in EU procurement law (Directive 2014/24/EU, 2014). Adjustments to these bands must be justified ex ante based on complexity, risk asymmetry, or market structure (DFØ, 2023). The LE rubric should reward verifiable track records and demonstrable project experience, while the RA rubric should prioritise precise risk allocations and mitigation strategies (Tran & Molenaar, 2016). Interview assessments must test not only narrative consistency but also the candidate’s ability to quantify impacts in KPI terms and articulate escalation pathways (D. Kashiwagi, 2016; Nguyen et al., 2018).

7.4. Formalising the VA Annex

The VA annex should comprise a single-page document detailing proposed improvements. These must include a safety justification, quantified effects on KPIs (without altering the base price), clear identification of the risk owner, verification mechanisms, and any preconditions such as permitting or sequencing (Storteboom et al., 2017; PIANOo/Rijkswaterstaat, 2023). Proposals that meet a clear ten-point screen—covering feasibility, measurability, and public-interest alignment—should be contractually embedded during Clarification to ensure accountability without compromising legal defensibility.

7.5. Codifying Information Instruments

To enable exception-based oversight, WRRs should be restricted to one page, tracking key risk metrics such as probability, impact, and trend, along with mitigation actions and due dates. DRs should summarise schedule deviations, cost variances, critical risks, and any decisions required by the contracting authority. A concise KPI set should monitor dimensions such as defect density, change-order rate, milestone adherence, and client decision load (D. Kashiwagi, 2016; DFØ, 2023). Central standardisation of templates enhances comparability and auditability (PIANOo/Rijkswaterstaat, 2023).

7.6. Roles and Responsibilities

Accountability remains with the contracting authority throughout the procurement lifecycle, but specific responsibilities vary by phase. The evaluation panel oversees Selection, while the BVA coach supports Clarification and maintains oversight during Execution. The vendor is responsible for WRR/DR compliance, and legal or audit functions should be integrated to ensure compliance and traceability (Directive 2014/24/EU, 2014; DFØ, 2023).

7.7. Legal Defensibility

To maintain legal compliance within Czech and EU procurement frameworks, evaluation criteria, rubrics, and weightings must be disclosed prior to bid submission. Scope and price must be locked during Clarification, and post-award price renegotiations should be disallowed (Directive 2014/24/EU, 2014). All documentation, including WRRs, DRs, and KPI reports, must be archived and version-controlled to ensure auditability (European Court of Auditors, 2023).

7.8. Monitoring and Organisational Learning

Agencies should track the performance of pilot tenders against historical baselines. Key indicators include procurement cycle time, staff effort, change-order rates, and delivery variance. Quarterly reviews can then adjust the scoring thresholds, VA criteria, and KPI definitions. A final fourth-quarter memo should consolidate lessons learned, refine templates, and authorise further scale-up (Beach & Pedersen, 2019; DFØ, 2023; PIANOo/Rijkswaterstaat, 2023).

8. Limitations and Future Research

8.1. Limitations

This study examined a single above-threshold construction tender supported by unusually rich documentary evidence. While this depth of material allows for fine-grained tracing of governance mechanisms, the scope of analytical generalisation is necessarily constrained. The findings speak most directly to price-dominant, legality-oriented administrative systems similar to the Czech context and should not be extrapolated to procurement regimes with substantially different institutional incentives or evaluative cultures without further empirical corroboration.

A second limitation concerns partial observability. Although WRRs, DRs, KPI updates and Clarification documents provide substantial insight into performance governance, access to the complete risk register was restricted by confidentiality provisions under Czech procurement law. Certain contractor-internal assessments and commercially sensitive risk attributions were unavailable. Triangulation across accessible artefacts mitigates—but cannot eliminate—the interpretive gaps created by these redactions.

Third, the case reflects a constrained rather than canonical configuration of BVA. The 40% price weight, omission of the VA mechanism, and reliance on introductory rather than continuous facilitation represent real-world institutional conditions but limit the method’s theoretical amplitude. As a result, the observed governance effects likely underestimate the potential performance outcomes achievable under mature enabling conditions.

Finally, the analysis does not rely on a randomised or quasi-experimental counterfactual (Shadish et al., 2002). Comparative insights are derived from historical project documentation and international benchmarks, which restrict causal identification. Although directed content analysis and mechanism-based reasoning strengthen internal validity, residual threats to inference remain, including unobserved confounders, selection effects, and contextual heterogeneity. The conclusions presented here should therefore be interpreted as analytical, rather than causal or statistical, generalisations.

8.2. Future Research

Future research can extend the insights of this study in several directions (Abadie et al., 2010).

First, cross-sector BVA pilots—for example in digital services, infrastructure maintenance, or complex public service contracts—would allow researchers to test the portability of weighting schemes, VA procedures, and facilitation models across different market and risk environments (Callaway & Sant’Anna, 2021).

Second, mechanism-oriented audits that map decision rights, information flows, and escalation behaviour could help quantify how BVA reallocates client decision load and where fidelity drift arises. Linking WRR/DR logs with KPI time series would enable deeper analysis of variance-driven oversight and risk-closure dynamics.

Third, quasi-experimental designs could improve causal identification. Potential approaches include matched-tender comparisons, difference-in-differences across staggered BVA adoption, or synthetic control methods to approximate counterfactual performance trajectories. Such designs could estimate impacts on evaluation effort, delivery variance, transaction time, and change-order incidence where pre-intervention trends are stable and assignment timing plausibly exogenous.

Fourth, policy evaluations of institutional complements—such as recommended price-weight ranges, structured VA annexes, or facilitation requirements—would clarify which enabling conditions are most influential for achieving lifecycle value creation. Creating shared repositories of KPI definitions, templates, and replication materials would support cumulative learning across authorities.

Finally, future work should assess the effects of varying facilitation intensity (Angrist & Pischke, 2009). Comparative studies of tenders receiving comprehensive coaching versus those receiving only introductory training would illuminate how facilitation shapes capability sorting, interpretive coherence, escalation quality, and risk-closure rates. Such research would contribute to theory-building at the intersection of procurement governance, institutional design, and organisational capability.