Scenario 06 — External Partner Collaboration

Part 01 · Scenario boundaries

What this page covers, and what it does not

Move the conversation from "evaluating a vendor" to "governing a cross-organisational mechanism."

Scene description

Scenario 02 determines which programs and milestones deserve priority. Scenario 03 identifies the constraint around which resources should be organised. Scenario 05 shows how a cross-functional issue becomes a clear decision. Scenario 06 builds on those foundations and asks:

When CROs, SMOs, central laboratories, imaging providers, bioanalytical laboratories, technology platforms, logistics providers or other service providers perform critical trial activities, how does a Clinical Operations leader ensure that external delivery serves the trial rather than stopping at contracted tasks and status reporting?

External collaboration differs from managing an internal team:

【1】The parties do not sit within one administrative hierarchy.

【2】Day-to-day delivery information is naturally distributed.

【3】Trial objectives, commercial objectives and risk preferences do not fully coincide.

【4】Clinical trials involve changes that cannot be predicted accurately at contract signature.

【5】A lead provider may subcontract further, extending the accountability chain.

【6】Activities can be transferred; related oversight and ultimate responsibility may not be transferable.

Mature partner governance cannot depend on the contract alone or on the relationship alone. Contract-only governance tends to become an exercise in assigning blame after a deviation. Relationship-only governance tends to depend on personal commitments and informal workarounds. Effective collaboration combines documented expectations, operational evidence, reciprocal commitments, proportionate incentives and continuing feedback.

Scope

External partners in this page may include: CROs; SMOs and site-support organisations; central laboratories, imaging providers and bioanalytical laboratories; EDC, IRT, eCOA, DCT and other technology providers; clinical supply, depot, cold-chain and logistics providers; recruitment, translation, printing and document-management providers; subcontractors engaged by a lead service provider.

A Clinical Operations leader may select and oversee a primary service provider, or may work within a CRO or another operations organisation and govern subcontractors, regional partners and site-support networks. The judgement model does not assume a single organisational viewpoint.

This page covers

Why do delegated relationships naturally create goal differences and information asymmetry?
What is each party optimising, and what costs are being transferred elsewhere?
How should controllable behaviour, shared outcomes and external influences be separated?
Why can a simple performance bonus create new metric distortions?
How can reciprocal commitments replace one-sided vendor accountability?
Which non-financial incentives remain practical when the trial budget is constrained?
How can risk transparency be protected from performance pressure?
How should feedback be used to test whether an incentive mechanism is improving the trial?
How can partner experience improve future selection and organisational capability?

Out of scope

The full procurement and competitive-bidding process
Professional legal advice on contract terms
Technical audit checklists for a specific vendor category
Comparisons of named CROs or service providers
Equal oversight intensity for every provider
Replacing quality oversight and professional judgement with bonus mechanisms
Interpreting ordinary commercial disagreement as an ethical failure
The full method for CAPA investigation and effectiveness review

Expected outputs

External-partner segmentation and governance-intensity map
Delegated-activity, retained-responsibility and critical-interface register
Principal-agent risk assessment
Bilateral externality map
Project conditional payment clause
Conditional-payment design review
Reciprocal project commitments
Provider issue diagnosis and escalation record
Relationship review and learning-transfer record

Part 02 · Breaking down the problem

Core question

Within budget, contract and organisational constraints, how can a Clinical Operations leader make the day-to-day choices of both parties more consistent with the outcome the trial actually needs?

Six judgements around the core question

1. Are the parties pursuing the same result?

What trial result is the commissioning organisation trying to protect?
What are the provider's contractual objectives, margin expectations and internal performance measures?
Which activities matter to the trial but are absent from the contract or scorecard?
Which metrics have become the team's real priorities?
Who benefits when a task is completed on time?
If the overall trial still fails, does task completion remain advantageous to one party?

2. What can each party actually see?

Who is closest to daily execution?
Who holds portfolio context, development strategy and internal decision information?
Can reported status be traced to reliable evidence?
Are rework, waiting, failed attempts and unresolved dependencies visible in governance?
Is the performance of critical subcontractors visible?
Which adverse updates may be delayed because of commercial or performance pressure?
Does the trial organisation explain changes in requirements and priority soon enough?

3. Who receives the benefit, and who bears the cost?

Who saves when training, review or senior staffing is reduced?
Who carries the later rework, site delay or quality risk?
How much waiting and resource conflict does the external team absorb when internal decisions are slow?
Are the costs of frequent requirement changes identified and confirmed promptly?
Does one function improve its own delivery by pushing problems into another team or vendor?
Is a short-term saving creating a larger opportunity cost for the trial?
Are participants, sites or downstream data users bearing consequences that neither contracting party bears directly?

4. Which behaviours are genuinely controllable?

How many parties influence the milestone?
Which actions are directly controlled by the CRO or service provider?
Which conditions must be supplied by the commissioning organisation?
Which outcomes depend on sites, participants, regulators or the external environment?
Is the vendor being evaluated on outcomes outside reasonable control?
Are internal decision and input responsibilities hidden behind vendor KPIs?
Is the mechanism measuring effort, process, outcome, or a balanced combination?

5. What behaviour does the current mechanism reward?

Can the bonus metric be optimised or cosmetically improved?
Does a penalty make teams slower to report risk?
Does visit volume displace risk assessment and meaningful site support?
Does query-closure speed weaken the quality of resolution?
Are future work opportunities governed by transparent criteria?
Does strong performance genuinely affect resources, discretion or scope?
Does the governance model distinguish early disclosure from concealment?

6. How can the right behaviour become the more rational choice?

Can clearer requirements and faster decisions reduce delivery cost?
Can a mature team receive bounded discretion?
Can resource forecasts and future opportunities become more transparent?
Can joint training and shared tools reduce delivery difficulty?
Is a limited conditional-payment mechanism appropriate?
Does the bonus include guardrails for quality, transparency and controllability?
What evidence will show that the mechanism is improving the trial?
When should the mechanism be adjusted, stopped or escalated?

Part 03 · Judgement framework

COD judgement framework: from delegated work to aligned action

External collaborations rarely fail because no contract exists. More often: the contract defines tasks but not the shared outcome; one organisation owns execution detail while another owns decision context; a saving appears in one budget while rework and risk appear elsewhere; performance measures reward local completion without protecting integrated delivery; once performance deteriorates, the parties debate responsibility instead of changing the conditions that produced the problem.

Scenario 06 uses a four-part framework:

Expose goal and information gaps → trace externalised costs → design reciprocal incentives → recalibrate through feedback

This framework is distinct from Scenario 05. Scenario 05 clarifies who provides input, recommends, decides and performs. Scenario 06 examines why independent organisations make different choices and how the mechanism behind those choices can be changed.

Four-step framework from delegated work to aligned action — **Figure 1.** From delegated work to aligned action — four-step path: ① goal and information gaps → ② externalised costs → ③ reciprocal incentives → ④ feedback recalibration.

1. Expose goal and information gaps

Find what does not align automatically within the contractual relationship

Do not begin with the number of tasks a provider must complete. Ask: which trial outcome does this relationship support? Which participant-protection, data-quality and milestone conditions cannot be lost? How does the provider's delivery enter the integrated trial path? Which tasks could be completed on time while the trial still fails? For example, the shared outcome of outsourced site start-up is not "collect the documents on time." The outcome is that appropriate sites become genuinely ready to recruit within the target window. Document collection is one activity within that result.

The commissioning organisation may focus on: the integrated trial timeline; a critical development window; quality and compliance risk; use of budget; cross-program resources and organisational reputation. The service provider may focus on: delivery within contracted scope; project revenue and margin; staff utilisation; internal performance measures; confirmation of changes and out-of-scope work; the client relationship and future business. The existence of different objectives does not imply that either party is unprofessional. The relevant question is where those differences change resource allocation, risk reporting or issue response.

The trial team should determine: which delivery information is held by the service provider; which decisions and strategic context are held by the commissioning organisation; what each party needs in order to make a sound choice; how evidence enters governance; which information does not need to be shared; how appropriate visibility is maintained over subcontracted activity. Transparency does not require unrestricted access to all internal provider information. Governance needs the evidence required to assess whether important activities remain healthy, risks are changing and controls are working.

The governance model should also specify: activities formally transferred; responsibilities retained; inputs and prerequisites; delivery standards and completion evidence; approval and decision paths; subcontracting arrangements; change control; risk and incident reporting; ownership of cross-organisational interfaces.

Outputs: shared-outcome statement, objective-gap analysis, information-asymmetry map, delegated and retained responsibility table, critical-interface map.

2. Trace externalised costs

Reveal the true trial cost of local optimisation

An externality occurs when a choice creates a cost or benefit for another participant that is not fully reflected in the decision of the party making that choice.

Costs transferred by a service provider: using less-experienced staff to reduce delivery cost while increasing review and rework for the trial team; reducing training hours while increasing site deviations and data issues; completing only in-scope tasks while leaving cross-scope dependencies unmanaged; delaying a revised forecast to protect short-term performance while compressing recovery time; subcontracting activity without bringing critical risk into joint governance.

Costs transferred by the trial organisation: incomplete requirements that force repeated estimates and rework; slow internal decisions that leave external resources waiting; frequent priority changes without adjusting time or budget; conflicting instructions from several functions; urgent work that begins before scope and fees are confirmed and remains unresolved; accountability for results without the information or discretion required to deliver.

For each externalised cost, ask: who made the choice? Who bears the consequence? Who has the authority to change the underlying condition? Why did the contract or scorecard fail to prevent the problem? Must the response change price, authority, information, workflow or future expectations? The purpose is not to calculate a perfectly precise transfer price. The purpose is to bring hidden costs, spread across budgets, organisations and time, back into the trial decision.

Outputs: bilateral externality map, cost bearer, controllable condition, required governance or commercial change.

3. Design reciprocal incentives

Make behaviour that protects the trial more compatible with each party's legitimate interests

An incentive is not limited to money. Any arrangement that changes the cost, benefit, authority or future value of a choice can influence behaviour.

Information incentives: governance should distinguish among risks found and reported early; deviations caused by capability or capacity; known problems left unreported; repeated failure to meet an explicit responsibility; unreliable evidence or misleading status. Early risk disclosure should lead to a timely decision, appropriate support and joint action. Concealment, distortion and late reporting should remain subject to formal accountability. Negative incentives should not attach automatically to reporting a safety or quality risk, honestly revising a forecast, identifying an infeasible protocol requirement or process, or raising a credible early warning before the evidence is complete.

Decision reciprocity: a service provider may commit to credible forecasts supported by evidence; stability in key roles; timely escalation of major dependencies and risks; a resource-recovery proposal when delivery is threatened; management of its subcontractors and internal interfaces; accountability for clearly defined delivery. The commissioning organisation should also commit to complete and reasonably stable requirements; decisions within agreed timeframes; control of unmanaged scope change; timely confirmation of changes, budget and priorities; coordination of internal functional input; no automatic penalty for transparent reporting of adverse news. Accountability without timely decisions, clear requirements or necessary authority is not incentive alignment. It is risk transfer.

Discretion as an incentive: the governance model can define issues the provider may resolve directly; resource changes permitted within an agreed boundary; circumstances in which risk can be contained before notification; decisions requiring prior approval; events requiring immediate escalation; evidence retained after an autonomous decision. Bounded discretion reduces waiting and allows a capable provider team to own an outcome rather than wait for instruction. The level of discretion should reflect capability, risk and performance.

Resource and future-opportunity incentives: in addition to current fees, external partners often value opportunities for future work; expansion of scope; visibility into future pipeline demand; more stable resource forecasting; a longer-term position in a therapeutic area; recognition of strong teams at governance level; opportunities to develop methods, tools and capability jointly. Future opportunities should not become vague promises. Criteria may include delivery reliability, stability of key roles, risk transparency, quality performance, recovery capability and contribution to cross-program learning.

Joint capability development: some delivery failures reflect missing context, weak interfaces or immature tools rather than a lack of effort. Practical responses include sharing necessary program and study context earlier; involving critical providers in protocol-operability assessment; joint training for critical processes; access to relevant historical risk examples; shared playbooks and issue libraries; earlier involvement of provider specialists in operational design; transfer of strong practices across programs. Capability development reduces delivery difficulty for the provider and future review and rework for the trial team.

Limited and balanced financial incentives: conditional incentive and penalty clauses can remain part of the design, but they should not become the entire mechanism. A common structure defines three performance levels: below the basic standard, reduced payment, recovery requirement or another agreed consequence; at the basic standard, normal contracted payment; materially above the standard, a limited additional payment. The design should meet several conditions.

【1】Measures are linked to behaviour within the provider's reasonable control.

【2】Outcome measures include quality and risk guardrails.

【3】A single volume measure does not determine the full reward.

【4】Early risk reporting does not automatically trigger a deduction.

【5】Waiting and scope change caused by the commissioning organisation are recorded separately.

【6】Caps, assessment periods and evidence sources are explicit.

【7】The reward can influence the team that allocates resources and performs the work.

【8】Underperformance is separated into capability, capacity, interface and external causes.

If the payment does not cover the marginal cost of additional resources, the bonus is unlikely to change resource allocation. A change in scope, priority, authority, future opportunity or delivery model may be more effective.

Anti-gaming guardrails: every incentive mechanism should be tested. Can the metric be split, delayed or reclassified? Will the team sacrifice first-time quality for speed? Will an average hide a high-risk site or process? Will status remain green until the last possible moment? Are leading, process and outcome signals reviewed together? Who verifies critical evidence independently?

Outputs: reciprocal commitment table, non-financial incentive portfolio, conditional-payment design, quality and transparency guardrails, anti-gaming review.

4. Recalibrate through feedback

Determine whether the mechanism changed real behaviour and trial outcomes

An incentive written into a contract or governance charter is not necessarily effective. Continuing feedback should test whether risks are raised earlier; whether forecasts are more credible; whether critical roles are more stable; whether rework and waiting have fallen; whether both parties make decisions more promptly; whether quality and milestones have improved; whether teams are optimising only the rewarded measure; whether the commercial model has created defensive behaviour.

Use four types of evidence together:

Evidence type	Primary question	Example
SLA	Was the contracted service performed within the agreed standard?	Ticket response, report submission, sample-transport time
KPI	Is critical work progressing as planned?	Site start-up, monitoring and data-cleaning progress
Key quality indicator	Does delivery protect the critical-to-quality factors?	First-time quality, important deviation trends, critical-data completeness
Risk and behavioural signal	Are new risks or healthy behaviours emerging?	Turnover, rework, forecast revisions, early escalation, waiting time

This page treats KQIs as feedback evidence, not as a universal regulatory list. Measures should be derived from trial CTQ factors, major failure modes and governance needs.

Each material incentive should define: the behaviour to be changed; observable evidence; data source; review point; expected improvement; possible unintended effects; adjustment or stop criteria; owner of follow-up. The review should inform future provider selection; SOW and contract templates; quality agreements; governance design; metrics and dashboards; trial start-up and handover; the vendor knowledge base; the next segmentation and scope decision.

Outputs: incentive-effectiveness review, mechanism-adjustment decision, partner-segmentation change, selection and contract learning record.

Mechanism calibration closed loop — **Figure 2.** Framework closed loop — principal-agent theory locates the misalignment, externality analysis exposes the true cost, incentive compatibility changes the choice, feedback tests the result. A mature collaboration is judged by what reaches governance while action is still possible, by whether both parties remain accountable for behaviour within their control, by whether a local saving is recognised when it creates a larger trial cost, by whether strong resources and constructive behaviour receive a credible return, and by whether an ineffective mechanism can be changed promptly when evidence requires it.

Part 03 · Principal-agent theory

Why goals do not align automatically once a contract is signed

Principal-agent theory examines how differences in goals, information and monitoring cost affect a relationship in which one party performs work for another.

Three issues

Adverse selection: during selection, an organisation can review corporate capability, standard processes and prior experience. It may still be difficult to see the team that will actually work on the trial; the experience of key roles; regional capacity; management span; competing workload; subcontracting arrangements; recovery capability under pressure. The response is not an unlimited questionnaire. Assessment should reach the real team and delivery model, followed by an early fit review during execution.

Moral hazard: the term is an economic concept, not a moral accusation. One party cannot fully observe the action of another, while the cost of effort and the consequence of outcomes are borne differently. A provider may submit a report on time while reducing the depth of investigation. A team may protect in-scope measures while leaving cross-scope dependencies unmanaged. Additional work may also be delayed when commercial confirmation remains unresolved. Responses include visibility of critical evidence; balanced observation of process, outcome and quality; timely change control; separation of early disclosure from concealment; formal accountability for repeated distortion or non-performance.

Multiple principals and goal substitution: a CRO team may respond simultaneously to client governance, internal functions, regional management, the quality system and subcontractors. When requirements conflict, the operational team prioritises according to authority, scorecards and commercial risk. If the performance model focuses only on visit volume, reporting time, ticket closure or query counts, teams will optimise those measures. Important work outside the scorecard may receive less attention.

The main lesson is not "do not trust the provider." It is: do not assume that goals, information and action align automatically once the contract is signed.

Part 03 · Externalities

Why one party's saving can become the real cost of the trial

Externality analysis brings costs dispersed across organisations, budgets and time back into one trial view.

Bilateral externality examples

Less senior staffing lowers commercial cost while increasing review and rework; fewer training hours reduce current effort while increasing site and data issues; delayed internal approval has little visible cost for the approving function while external resources wait and the development window narrows; repeated requirement changes preserve flexibility for one team while the provider absorbs reconfiguration and planning failure; one vendor protects its own SLA while the trial carries the risk at the vendor interface.

Externalities are bilateral and often involve several parties. The Clinical Operations leader should ask: what uncontrollable cost is the trial organisation creating for the external team? What consequences are sites and participants bearing? Is local functional efficiency reducing system-level efficiency? Which present saving is being shifted into a future milestone or quality risk?

This analysis moves incentive design beyond "how much was completed, and how much should be paid?" It identifies the true source of cost.

Part 03 · Incentive compatibility

Why an advanced mechanism is not centred on bonuses

Incentive compatibility comes from mechanism design. The question is not how to persuade a party to cooperate. The question is how rules can make choices consistent with the shared objective while participants still protect their legitimate interests.

The reality of a multitask environment

Clinical trials are a classic multitask environment: many outcomes are influenced jointly; important work can be difficult to measure; quality and risk consequences emerge with a delay; stronger rewards for one measure can pull effort away from other tasks; strong penalties can reduce transparency; a commercial reward may not reach the operational team.

The mechanism therefore needs a combination of clear outcomes; controllable accountability; information transparency; reciprocal commitments; bounded discretion; resources and future opportunities; joint capability development; limited financial incentives; quality and risk guardrails; feedback and adjustment. A conditional bonus can be one component. It cannot substitute for the full mechanism.

Part 03 · Supporting theory

Relational contracting: why a contract alone is not enough

Clinical trials involve change that cannot be forecast precisely at contract signature. A formal agreement can define scope, price, responsibility and remedies. It cannot specify every future circumstance.

What relational contracting adds

Relational-contract theory highlights the importance of shared purpose; information transparency; reciprocity; integrity of roles; adaptive response; conflict resolution; the future value of the relationship.

Relational governance does not replace the written contract. Mature collaboration requires both: the contract provides boundaries, rights, responsibilities and enforceability; relational governance supports adaptation, coordination, trust and long-term adjustment.

Part 04 · Worked example

Designing CRO site-start-up incentives under a constrained budget

The example below illustrates a method. It does not describe a specific trial or service provider, and it should not be read as a universal governance model.

Starting point

A CRO performs site identification, start-up management and monitoring activities for a multicentre trial. Start-up is behind plan, and the parties begin to discuss conditional incentive and penalty clauses. The first proposal is simple: below the agreed start-up standard, reduce part of the payment; at the standard, pay the contracted amount; above the standard, pay an additional bonus.

The proposal appears outcome-oriented. Further review reveals several problems: site activation depends on contracts, ethics review, internal approvals, systems, laboratories and supply readiness; the CRO controls some activities but not all conditions; rewarding activation volume can favour easy-to-activate sites with weak recruitment potential; a penalty may make the team slower to disclose deterioration in forecast dates; the additional payment may not reach the staff who perform the start-up work; the trial budget cannot purchase increasing resources indefinitely.

The team retains a financial component but redesigns the wider mechanism.

Define the shared outcome

The outcome is no longer "activate more sites." It becomes: bring sites with credible recruitment potential to genuine recruitment readiness within the target window, without trading away start-up quality, and improve the credibility of start-up forecasts. The result has four layers.

【1】Site quality: selected sites fit the participant pathway and recruitment strategy.

【2】Readiness: contract, ethics, documents, systems, laboratory, supply and staff training are genuinely ready.

【3】Time performance: critical-path waiting and rework are controlled.

【4】Information quality: date forecasts and risk reporting support trial decisions.

Identify bilateral externalities

Potential CRO externalities: less-experienced start-up staff create rework; activation volume is prioritised over true site potential; forecast deterioration is reported late, compressing recovery time; laboratory, system and other vendor dependencies are not integrated.

Potential trial-organisation externalities: slow site and budget decisions; continuing changes in protocol or start-up requirements; inconsistent instructions from several functions; work begins before scope and fees are confirmed; an accelerated timeline is requested without faster decisions or a revised site strategy.

The parties agree that start-up cannot be treated solely as a CRO score, nor can every result be attributed to internal dependency.

Establish reciprocal commitments

CRO commitments: provide a site-level critical path supported by evidence; explain each material revision to an activation forecast; keep critical start-up roles stable; propose a resource-recovery plan for high-potential sites; raise cross-vendor dependencies early; remain accountable for first-time submission quality and rework causes.

Trial-organisation commitments: make site and budget decisions within agreed timeframes; provide complete and reasonably stable start-up requirements; confirm changes in scope and priority promptly; assign a decision owner and escalation path for cross-functional blockers; avoid penalising credible early risk reporting; record and own internal waiting time.

Combine non-financial incentives

Resource visibility: the CRO receives earlier information about future phases and potential workload, making retention of a mature team easier.

Preferred future opportunities: risk transparency, stability of key roles, recovery capability and cross-program learning enter future scope decisions. Strong performance can improve access to future work, but no unconditional award is promised.

Bounded discretion: within agreed limits, the CRO may adjust start-up staffing, change support intensity and resolve low-risk dependencies without waiting for case-by-case approval.

Faster decisions: high-potential-site budget, contract and exception decisions receive explicit internal turnaround times.

Joint capability development: the parties review the causes of start-up-package rework and improve templates, training and first-time submission checks.

Governance visibility: strong provider teams receive explicit recognition in joint governance, and team performance becomes part of future relationship review.

Retain a limited financial component

Financial incentives no longer depend only on activation volume.

Basic guardrails: speed is not achieved at the expense of quality or risk transparency; critical start-up evidence is complete; material dependencies are escalated promptly; high-potential-site priority is followed.

Performance dimensions: timely completion of activities within the provider's control; first-time submission quality; accuracy of activation forecasts; reduction in critical waiting and rework; management of cross-party dependencies within CRO scope; time from activation to genuine screening readiness.

Protection conditions: delays caused by internal decisions, requirement changes or uncontrollable external factors are treated separately; early risk reporting does not automatically trigger a penalty; volume performance must meet quality thresholds; the reward is capped; assessment can be traced to reliable evidence; the path through which commercial reward reaches the project team is agreed.

Test through feedback

After an agreed period, the team reviews more than activation volume: are risks visible earlier? Are date forecasts more credible? Has first-time quality improved? Has waiting on both sides fallen? Are high-potential sites receiving concentrated support? Has the CRO stabilised critical roles? Is any new form of metric gaming visible? Did the payment change resource allocation?

If the bonus does not change the resource decision, the parties should adjust scope, priority or staffing rather than continue to increase an ineffective reward.

The example demonstrates the role of the Clinical Operations leader in Scenario 06: not merely scoring a vendor, but exposing differences in objectives, information and costs, then designing a practical mechanism that changes day-to-day choices.

Part 05 · Critical actions

Critical actions for the Clinical Operations leader

Define the external-partner outcome from trial results and CTQ factors.
Confirm delegated activities, retained responsibilities, subcontracting and critical interfaces.
Identify differences in objectives, information and internal performance systems.
Determine whether a local saving is creating an external cost for the trial.
Separate controllable provider behaviour, shared outcomes and external factors.
Design reciprocal commitments rather than vendor-only KPIs.
Use information, discretion, resources, future opportunities and capability development before relying on cash.
Attach quality, transparency, controllability and anti-gaming guardrails to financial incentives.
Use feedback to test whether the mechanism changed real behaviour.
Return the learning to selection, contracting, governance and organisational knowledge.

Part 05 · Role boundaries

Boundaries the Clinical Operations leader should not cross

Coordination responsibility does not mean replacing professional functions.

Function boundaries at a glance

Procurement and vendor-management functions: own procurement processes, commercial negotiation, vendor policy and commercial contract management. The Clinical Operations leader provides trial requirements, performance evidence and business judgement without replacing professional procurement responsibilities.

Legal: owns legal risk, liability clauses, disputes and remedies. Clinical Operations should provide clear operational facts, trial impact and intended outcome rather than legal conclusions.

Quality Assurance: provides independent quality judgement, audit and oversight under the quality system. Clinical Operations should not present an operational-performance view as a formal Quality conclusion or replace independent investigation.

Clinical Development, Data, Safety, Biostatistics, Regulatory and other functions: own professional standards and judgement within their domains. When an external provider performs related activities, the relevant function still provides appropriate input, review and oversight.

Service-provider project leadership: owns the provider's team, resources and internal management. The Clinical Operations leader may require outcomes, evidence and a recovery proposal, but should not routinely bypass provider leadership to micromanage individuals.

Clinical Operations leadership: brings the trial objective, operational evidence, external delivery, reciprocal dependencies and risk into one governance view. Clinical Operations ensures that boundaries are clear, important information is visible, incentives are proportionate and the collaboration continues to support the trial.

Related capabilities

External-partner selection and continuing-fit assessment; scope and delegated-responsibility design; service-provider oversight; commercial understanding and contract awareness; principal-agent risk assessment; identification of externalised cost; incentive-mechanism design; metric design and evidence judgement; reciprocal commitment and difficult conversations; interface management; risk escalation; partner segmentation and relationship governance; organisational learning.

Part 06 · Project tools

Tools that can enter the project workspace

1. CRO principal-agent risk assessment

Use when designing an outsourcing model, starting a trial, reviewing persistent underperformance or resolving different interpretations of responsibility. Core fields: shared trial outcome; commissioning-organisation objective; provider objective; objective gap; information asymmetry; unobservable effort; controllable behaviour; current oversight; current incentive; potential metric distortion; proposed change.

2. Bilateral externality map

Use when one party attributes delay or cost to the other but the full delivery chain has not been reconstructed. Core fields: choice or behaviour; party making the choice; local benefit; party bearing the cost; timeline, quality, budget and resource effect; presence in contract or scorecard; authority to change the condition; proposed mechanism.

3. Project conditional payment clause

Use when translating a critical project outcome, three-tier payment conditions and necessary safeguards into a work order, amendment or project-level commercial arrangement. Core fields: applicable project stage, milestone or deliverable; basic delivery standard; criteria for below-standard, at-standard and above-standard performance; payment, deduction, recovery requirement or other commercial treatment at each tier; behaviour within the service provider's reasonable control; inputs and decisions the trial organisation must provide on time; exclusions for protocol change, internal waiting or uncontrollable external events; participant-protection, data-reliability and critical-quality thresholds; protection for early risk disclosure; assessment period, evidence source and evidence approver; caps on additional payment and deduction; dispute handling, review point and clause termination criteria. Use note: the clause template makes an agreed project mechanism concrete. It does not replace professional review by Procurement, Legal or Quality.

4. Conditional-payment design checklist

Use when considering a performance payment, bonus, deduction or risk-sharing arrangement. Questions: is the measure controllable? Is the baseline credible? Is there a quality threshold? Is early risk disclosure protected? Is waiting caused by both parties recorded? Can the metric be gamed? Can the reward reach the delivery team? Is the reward large enough to change resource allocation? Are the assessment period and evidence source clear? Are there caps, review points and stop conditions?

5. Reciprocal project commitment table

Use when the parties debate accountability without clear commitments for inputs, decisions and delivery. Core fields: service-provider commitment; trial-organisation commitment; owner; required input; due date; completion evidence; dependency; effect of non-performance; escalation path.

6. Incentive-effectiveness review

Use when a new performance mechanism has operated long enough to warrant review. Core fields: intended behaviour; observed behaviour change; trial-result change; supporting evidence; unintended effect; sign of gaming; resource impact; feedback from both parties; continue, adjust or stop decision; learning to be transferred.

Part 06 · Intellectual and regulatory foundations

Foundations behind the methodology

The three tools answer different questions. Principal-agent theory locates the misalignment. Externality analysis exposes the true cost. Incentive compatibility changes the choice. Feedback tests the result.

Conceptual source

Once a delegated relationship exists, the parties still hold different objectives, information, costs and internal scorecards. Principal-agent theory helps the Clinical Operations leader locate those structural differences. Externality analysis returns costs scattered across organisations, budgets and time to the trial view. The analysis covers provider behaviour as well as late internal decisions, scope drift and interface failures. Incentive compatibility is broader than a bonus. Information, discretion, decision speed, future opportunity, capability support and limited financial consequences can all shape behaviour. Both parties should be accountable for controllable choices without sacrificing quality or risk transparency.

Regulatory foundation

ICH E6(R3) includes directly relevant expectations for service-provider governance:

【1】Trial activities transferred to and assumed by a service provider should be documented in an agreement.

【2】Activities not specifically transferred remain with the original responsible party.

【3】For sponsor activities, ultimate responsibility for participant rights, safety and well-being and for the reliability of trial data remains with the sponsor even when activities are transferred.

【4】Service providers should implement appropriate quality management and report incidents that may affect participant safety or trial results.

【5】The sponsor should assess the suitability of and select the service provider.

【6】Relevant information, such as SOPs and performance metrics, should be accessible for selection and oversight.

【7】Important transferred activities should receive appropriate oversight, including activity subcontracted further.

These requirements explain why appropriate judgement and oversight remain necessary after outsourcing. The guideline does not prescribe one bonus or performance-payment model. Any incentive arrangement remains subordinate to participant protection, data reliability, proportionate risk management and applicable law.

Representative thinkers

Michael C. Jensen & William H. Meckling

Their 1976 paper provided a foundational account of agency relationships and agency costs. Scenario 06 uses this perspective to examine goal differences, information asymmetry, unobservable effort and monitoring cost. Agency theory should not be reduced to the claim that "vendors will shirk."

Arthur C. Pigou & Ronald H. Coase

Pigou's work in welfare economics helped establish the systematic analysis of external effects. Coase later emphasised the reciprocal character of external harm and the importance of institutional arrangements, transaction costs and rights. Scenario 06 uses these perspectives to examine how both parties can transfer cost to one another and to the wider trial.

Leonid Hurwicz

A foundational figure in mechanism design and a key representative of incentive compatibility. Scenario 06 draws on the central mechanism-design question: not how to persuade a provider, but how rules shape the choices of participants.

Bengt Holmström & Paul Milgrom

Their work on multitask principal-agent problems shows how rewarding a limited set of measurable tasks can draw effort away from other important work. Scenario 06 uses that perspective to challenge reliance on a single volume or time measure in clinical-trial provider performance.

Ian R. Macneil

One of the principal thinkers associated with relational-contract theory. Scenario 06 uses relational contracting as supporting theory to explain why formal agreements need complementary norms of transparency, reciprocity, adaptation and continuing cooperation.

References

International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use. (2025). ICH E6(R3) Guideline for Good Clinical Practice: Principles and Annex 1, Sections 3.6.4–3.6.10.
European Parliament and Council of the European Union. (2014). Regulation (EU) No 536/2014 on Clinical Trials on Medicinal Products for Human Use, Article 71.
Jensen, M. C., & Meckling, W. H. (1976). Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure. Journal of Financial Economics, 3(4), 305–360.
Eisenhardt, K. M. (1989). Agency Theory: An Assessment and Review. Academy of Management Review, 14(1), 57–74.
Pigou, A. C. (1920). The Economics of Welfare. Macmillan.
Coase, R. H. (1960). The Problem of Social Cost. Journal of Law and Economics, 3, 1–44.
Hurwicz, L. (1972). On Informationally Decentralized Systems. In C. B. McGuire & R. Radner (Eds.), Decision and Organization. North-Holland.
Holmström, B., & Milgrom, P. (1991). Multitask Principal-Agent Analyses: Incentive Contracts, Asset Ownership, and Job Design. Journal of Law, Economics, & Organization, 7, 24–52.
Macneil, I. R. (1978). Contracts: Adjustment of Long-Term Economic Relations under Classical, Neoclassical, and Relational Contract Law. Northwestern University Law Review, 72, 854–905.
Baker, G., Gibbons, R., & Murphy, K. J. (2002). Relational Contracts and the Theory of the Firm. Quarterly Journal of Economics, 117(1), 39–84.

External partner collaboration