Complex national and organisational problems rarely fit inside one department.
Trade policy affects technology transfer. Technology standards affect investment. Data rules affect market access. Industrial capability affects national security and export competitiveness. Decisions in one institution create consequences that another must manage.
The common response is to create a committee, steering group, or interagency task force.
That may establish authority and representation. It does not automatically create coordinated decisions.
The deeper issue is the decision interface.
**A decision interface is the agreed structure through which an issue, evidence, authority, deadline, escalation, and execution status move across institutional boundaries.**
Vietnam's newly established interagency body for international economic and science-technology integration reflects the growing connection among trade agreements, foreign investment, technology transfer, supporting industries, export standards, strategic technology, supply chains, and the digital economy. The strategic value of this structure will depend on whether coordination becomes an operating capability rather than a calendar of meetings.
Representation does not guarantee resolution
A committee can include every relevant institution and still fail to resolve cross-boundary issues.
Typical failure patterns include:
- agencies arrive with different definitions of the problem;
- information is shared, but no decision is requested;
- members lack authority to commit their institutions;
- evidence is presented in incompatible formats;
- risks are transferred rather than jointly evaluated;
- action items are recorded without dependency owners;
- escalation occurs only after deadlines fail;
- implementation is reported as activity rather than verified outcome.
In these conditions, coordination absorbs attention without reducing ambiguity.
This is why [leaders need coordination truth](/blog/leaders-need-coordination-truth). Visibility becomes useful only when leaders can see where work is actually connected, blocked, or falsely assumed to be complete.
Cross-boundary work needs an interface
An interface is familiar in technology. Two systems can interact reliably when they agree on inputs, outputs, formats, permissions, and error handling.
Institutions need the same discipline.
A decision interface should answer:
- What issue is entering the system?
- Why can it not be resolved inside one institution?
- What decision is required?
- Which evidence must accompany it?
- Who has authority to decide?
- Which institutions must execute the result?
- What deadline reflects the real cost of delay?
- What happens when evidence or authority is insufficient?
- How will completion be verified?
Without these elements, each agency must reinterpret the issue at every handoff. Delay and distortion accumulate.
Build six layers of the decision interface
1. Entry criteria
Not every cross-functional issue belongs at the highest coordinating level.
Define what qualifies for interagency attention. Useful criteria may include statutory overlap, conflicting incentives, strategic technology, significant economic exposure, cross-border consequences, or a dependency that no single institution can resolve.
Clear entry criteria protect the mechanism from becoming a general discussion forum.
They also preserve subsidiarity: decisions should remain as close as possible to the people with relevant knowledge and execution capacity.
2. A common problem statement
Before debating solutions, participants need a shared description of the decision.
The problem statement should specify:
- the current condition;
- the affected actors;
- the consequence of inaction;
- the policy or operating constraint;
- the decision required;
- the deadline and why it matters.
This prevents one agency from discussing legal authority while another discusses budget, technology, or diplomatic risk without recognising that they are answering different questions.
3. A portable evidence packet
Evidence should arrive in a form that multiple institutions can use.
The packet may include relevant data, legal constraints, technical assessments, stakeholder impact, scenarios, dependencies, dissenting views, and unresolved uncertainty.
It should distinguish facts from assumptions and recommendations. It should also identify evidence that is missing but not realistically obtainable before a decision.
Portable evidence reduces repeated requests and protects against decisions based on whichever institution presents the most persuasive narrative.
4. Decision rights
Participation is not authority.
For each issue, clarify who can:
- recommend;
- approve;
- veto on defined grounds;
- request additional evidence;
- accept residual risk;
- assign execution;
- escalate unresolved conflict.
Decision rights may vary by topic. Trade negotiations, investment screening, technical standards, data rules, and research cooperation do not require identical authority structures.
This principle builds on [role contracts in performance leadership](/blog/performance-leadership-role-contracts). Accountability becomes fair and useful only when outcomes, authority, guardrails, evidence, and review are connected.
5. Time and escalation rules
Cross-boundary decisions often stall because delay has no owner.
The interface should define response windows, decision dates, escalation triggers, and temporary operating rules when full resolution is impossible.
Escalation should occur when:
- an institution cannot provide required evidence;
- authority is disputed;
- dependencies threaten the deadline;
- the risk exceeds the assigned decision level;
- implementation diverges from the agreed outcome.
Escalation is not a sign of failed coordination. It is a designed route for conditions the current level cannot resolve.
6. Execution verification
A signed decision does not guarantee a changed system.
The interface should carry the decision into implementation through named owners, dependency milestones, shared status, and evidence of the intended outcome.
Verification asks:
- Was the policy, standard, workflow, or agreement actually changed?
- Did connected institutions implement compatible changes?
- Can businesses, citizens, researchers, or partners use the result?
- Did the decision create a new bottleneck elsewhere?
- What should be revised after operating evidence appears?
This turns coordination into a learning loop.
Permanent interfaces, temporary teams
Institutions often create permanent committees for problems that require temporary concentrated work. The reverse can also occur: a temporary task force dissolves even though the coordination requirement is permanent.
The better design separates the interface from the team.
The interface can remain stable: entry rules, evidence format, decision rights, escalation, and verification. The people and expertise assembled for each issue can change.
This creates flexibility without losing institutional memory.
It also reduces meeting load. Participants join when their authority or expertise is required, not because they occupy a permanent seat in every conversation.
Use a dependency map, not an agency list
An agency list shows representation. A dependency map shows how results are produced.
For an international technology-investment issue, the map might connect:
- trade commitments;
- investment conditions;
- technology-transfer obligations;
- standards and certification;
- intellectual-property arrangements;
- domestic supplier capability;
- research partnerships;
- data and cybersecurity requirements;
- workforce development;
- diplomatic and commercial relationships.
The map reveals sequencing. A technology-transfer commitment creates limited value if local firms lack testing capacity. A new export opportunity may stall if certification rules are unresolved. Research cooperation may produce little value if intellectual-property and commercialisation rights remain ambiguous.
This is the same leadership logic behind [dependency mapping](/blog/leaders-need-dependency-mapping-not-only-succession-planning): leaders must see the system of reliance, not only the formal organisation chart.
Measure decision flow and implementation quality
Committee performance is often reported through meetings, documents, consultations, and attendance. These measures confirm activity, not coordination value.
More useful indicators include:
- time from issue entry to a clear decision request;
- percentage of cases with complete evidence packets;
- time waiting for authority or missing information;
- number of decisions reopened because dependencies were missed;
- implementation completion across all affected institutions;
- user or business friction removed;
- recurrence of the same unresolved cross-boundary issue;
- time from escalation trigger to higher-level decision;
- operating outcomes after implementation;
- lessons incorporated into the interface.
The objective is not always faster decisions. Some strategic choices require deliberation. The objective is purposeful time: leaders should know why a decision is waiting and what must happen next.
Leadership behaviour must support the interface
Even a well-designed process can fail when leaders defend jurisdiction instead of outcomes.
Effective interagency leadership requires several behaviours:
- state institutional constraints honestly;
- distinguish legitimate risk from territorial preference;
- expose dependencies early;
- send representatives with real authority;
- accept shared evidence standards;
- document dissent without paralysing action;
- protect teams that escalate genuine conflicts;
- review whether execution changed the intended system.
The most mature leaders do not seek to win every boundary dispute. They design a process in which the quality of the joint decision is stronger than any institution's isolated view.
Conclusion
Committees create a place for coordination. Decision interfaces create the conditions for coordinated action.
As economic policy, science, technology, investment, data, standards, and international relations become more interdependent, leadership cannot rely on representation and goodwill alone.
Issues need a defined entrance. Evidence must travel. Authority must be visible. Delay needs escalation. Decisions must reach implementation, and implementation must return evidence.
The strength of an interagency mechanism is not measured by how many institutions sit around the table. It is measured by how reliably the system converts shared complexity into accountable decisions and verified results.
Key Takeaways
- Cross-sector committees do not automatically create coordinated decisions.
- Decision interfaces define issue entry, evidence, authority, deadlines, escalation, and verification.
- A stable interface can support temporary teams assembled around specific problems.
- Dependency maps are more useful than lists of participating institutions.
- Performance should be measured through decision flow and implemented outcomes, not meeting volume.
FAQ
What is a decision interface?
It is the agreed structure through which an issue, evidence, authority, deadline, escalation, and execution status move across institutional boundaries.
Why do interagency committees often move slowly?
They may have representation without a shared problem statement, portable evidence, decision rights, time rules, or clear implementation ownership.
Should every cross-functional issue go to an interagency body?
No. Entry criteria should reserve the mechanism for issues that contain material dependencies or conflicts that one institution cannot resolve alone.
How should interagency performance be measured?
Measure decision latency, evidence quality, dependency resolution, escalation effectiveness, implementation across institutions, and the outcomes experienced by users or businesses.