Designing Effective Dispute Resolution Systems: A Practical Framework for Government and Enterprise Teams

Updated: 6 days ago

Designing effective dispute resolution systems means building the whole path a concern travels, not just writing a procedure. That includes accessible entry points, triage, several pathways matched to the problem, clear escalation and decision rights, trained people, and feedback loops that turn outcomes into improvements. Capability, not policy, determines whether the design works.
Most organisations I work with already have a complaints policy, a grievance procedure and a list of approved mediators. What many lack is a system: one connected design in which someone owns the early stage, the right matter reaches the right process, and the organisation learns from what happens.
In my experience, drawn from 15+ years of dispute resolution practice and work with 50+ government and business organisations (my own practice data), the same pattern repeats. Disputes rarely start as disputes. This article sets out a practical framework for heads of complaints, HR directors, L&D leads and directors who own or are reviewing how disputes flow through their organisation.
Key Takeaways
A process is a set of steps. A system is the full path from first concern to organisational learning, and it needs an owner at every stage, especially the early one.
Conflict is not a failure. It is often a signal that something important has not yet been properly understood, and a good system is built to hear it early.
Diagnose before you design. Your complaints register, review rates and staff workarounds usually show where the design is failing.
Not every matter needs the same process. Build pathways for coaching, facilitation, mediation, investigation and clear management decisions, then triage honestly between them.
Frontline staff and managers decide whether a system works. Design without capability building produces a document, not a system.
Feedback loops and regular review keep the design honest, and surface systemic issues that no single case can show.
Summary: The Six Parts of a Dispute Resolution System
Entry points
The question it answers: Where and how can someone raise a concern?
Typical failure: Only a formal channel exists, so people wait
What good looks like: Several low-barrier ways in, including to a person who can act
Triage
The question it answers: Which pathway fits this matter?
Typical failure: Default to the familiar tool
What good looks like: Clear criteria, documented reasoning, quick decisions
Multiple pathways
The question it answers: What are the options for resolution?
Typical failure: Mediation or investigation for everything
What good looks like: Coaching, facilitation, mediation, investigation and management decision, each with a defined use
Escalation and decision rights
The question it answers: Who can decide what, and when does it move up?
Typical failure: Unclear authority and invisible independence
What good looks like: Defined roles, fair review, independence where it matters
People capability
The question it answers: Can staff and managers actually use it?
Typical failure: Policy is written, skills are assumed
What good looks like: Shared language and practised skills across the cohort
Feedback loops
The question it answers: What is the system telling us?
Typical failure: Cases closed, patterns ignored
What good looks like: Themes reported, design reviewed on a schedule
Why Process Alone Is Not a System
A process is a set of steps for handling a matter. A system is the whole path a concern travels, from the moment someone first raises it to the point where the organisation learns from it. Most organisations have processes. Few have anyone who owns the early stage, where small issues are cheapest to resolve.
The issue is rarely just the original concern. Most workplace and service disputes do not begin with a dramatic event. They begin with a small issue that everyone decides to manage around: a decision that was not explained, a workload that quietly shifted, a reply that never came. By the time the matter surfaces formally, people are responding to the silence, the delay and the perceived unfairness as much as the original problem.
That is why I hold a position that runs against a common view. Conflict is not a failure, and it is not a sign of dysfunction to be eliminated. It is often a signal that something has not yet been properly understood. Treating it as shameful drives it underground, where it compounds. Treating it as information lets leaders respond early and in proportion.
What a late-stage-only design looks like
Consider a hypothetical regulator whose only formal option is a late-stage review. A person raises a concern with a frontline officer, who has no authority to resolve it and no guidance on what to do instead. The officer says it will be looked at. Weeks pass.
The person finally lodges a formal complaint, now angry about the delay as well as the original decision. The officer, feeling exposed, becomes defensive. The formal process must now deal with the first issue, the handling of the first issue, and a relationship that has soured.
Nothing in that story required bad intent. It required a design gap: no early pathway, no clear authority at the front line, and no one accountable for the first conversation. Policy documents cannot close that gap. A system can.
Two tests show whether you have a system or a collection of processes:
Can you name who owns the first 48 hours of a concern, and what they are permitted to do?
Can you show how outcomes from closed matters change how the organisation works?
If either answer is vague, the design is incomplete, however good the individual procedures are.
Diagnose Before You Design
Start by mapping what actually happens, not what the policy says should happen. Trace where disputes arise, who first receives them, which pathway they take, how long each stage lasts and where they stall. The data you already hold usually reveals the design problem before any consultation begins.

The temptation is to open a policy rewrite. Resist it. A rewritten policy on top of an undiagnosed system tends to reproduce the same failures in better language.
What the data already shows
Before you interview anyone, pull together what exists:
Complaints and grievance registers, including how each matter was categorised and closed.
Internal review, appeal and external referral volumes (for example to an ombudsman or tribunal).
Time spent at each stage, and time between the first contact and the first substantive response.
Repeat contacts from the same person or about the same issue.
Staff survey comments, exit interview themes and informal HR contacts that never became cases.
Outcomes: how many matters were resolved at the first point of contact, and how many arrived at formal stages already damaged.
Look for gaps as well as figures. If almost nothing is recorded at the early stage, that is not evidence of harmony. It usually means early concerns are being handled informally, invisibly and inconsistently.
Ask the people who manage around it
Data tells you where matters go. Conversation tells you why. Speak with frontline staff, team leaders, HR advisers, complaints officers and, where appropriate, people who have used the system. Useful questions include:
When someone raises a concern with you, what do you do first?
What stops you from resolving it yourself?
Which matters do you avoid referring on, and why?
When did you last see a matter that should have been settled earlier?
The workarounds are the diagnosis. Where staff route matters through an informal contact, a trusted manager or a quiet word with HR, the formal design has failed that moment, and you have found a pathway that needs to exist properly.
The work of Ury, Brett and Goldberg in Getting Disputes Resolved (1988) is a useful lens here. They argued that disputes can be resolved by reconciling interests, by determining rights or by testing power, and that well-designed systems push resolution towards interests wherever it is safe and appropriate to do so, keeping rights-based and power-based processes as a backstop.
Your diagnosis should ask how much of your current volume ends up in the rights-based stages purely because nothing earlier was available.
Build Multiple Pathways, Because Not Every Matter Needs Mediation
A sound system offers several pathways, each matched to a type of problem. Not every matter needs the same process. Some need coaching, some facilitation, some mediation, some investigation, and some simply a clear management decision. The hardest call is often which process to use, not what the outcome should be.

Mediation is valuable, and it is also overused as a default because it is familiar and sounds careful. Sending a conduct allegation to mediation can leave a serious matter unaddressed. Sending a communication breakdown between two capable colleagues to a formal investigation can create the very damage you were trying to prevent.
Conflict coaching
Fits when: One person needs skills, language or confidence to raise an issue or respond to one
Poor fit when: The matter involves safety risk or alleged misconduct that must be assessed
Facilitated conversation or group facilitation
Fits when: A team or group needs structure to resolve a shared issue, such as roles or workload
Poor fit when: The parties are not willing to engage, or the matter needs a decision rather than agreement
Mediation or conciliation
Fits when: Parties have a genuine dispute, can participate voluntarily and have authority to reach an outcome
Poor fit when: There is a serious power imbalance that cannot be managed, or a finding of fact is required
Investigation
Fits when: Facts are contested and may constitute misconduct or a breach of obligations
Poor fit when: The real problem is a relationship or process issue that no finding will fix
Clear management decision
Fits when: The matter turns on a decision that someone with authority must make and explain
Poor fit when: The decision has already been made unfairly or without explanation, and trust is the core problem
Use the Process-Fit Distinctions
When an organisation is about to default to a familiar tool, I use what I call the Process-Fit Distinctions. They are simple prompts:
Mediation vs investigation: Is the need to reach agreement, or to establish what happened?
Conflict vs misconduct: Is this a difference that can be worked through, or conduct that must be addressed?
Early resolution vs avoidance: Are we resolving this quickly, or hoping it goes away?
Neutrality vs fairness: Being neutral does not mean the process is fair to everyone in it.
Empathy vs agreement: Listening well is not the same as conceding.
Psychological safety vs comfort: Safety lets hard things be said. Comfort avoids them.
Give the early stage a structure
Pathways need a front door. For matters before formal escalation, I use the Early Resolution Sequence: clarify the issue, understand what matters, choose the right process, create structure, support the conversation, and document the next step. It turns a vague tension into a manageable process, and it is something frontline staff and managers can learn to use. Clarity is kind. Delay is not the same as care.
Triage, Escalation and Decision Rights
Triage is the point where a concern is matched to a pathway using stated criteria, by a defined role, within a set time. Escalation is the rule for when a matter moves up. Decision rights say who can decide what. Without all three, the system depends on individual judgement and goodwill.
Good triage is not a single person's instinct. It uses criteria that can be stated, tested and recorded. Useful criteria include:
Risk and safety, including psychosocial or physical risk to anyone involved.
Whether the matter may involve misconduct, discrimination or a legal obligation.
Power imbalance between the people involved.
Complexity and the number of people affected.
Urgency, including statutory or policy timeframes.
The willingness and capacity of the parties to take part in a joint process.
Any conflict of interest in who would normally handle the matter.
Record the reasoning for the pathway chosen, even briefly. It protects the person making the call, supports consistency and gives you data later.
Escalation should be a right, not a penalty
Write down when a matter must move up: when the time standard has passed, when a pathway has not resolved the issue, when new information changes the risk, or when someone requests review. People should not feel that escalating is an accusation. They should know the rule in advance, and so should the staff handling the matter.
Decision rights and visible independence
In government and regulated settings, procedural fairness shapes every stage. Define who can resolve a matter at the front line, who can approve remedies, who can change a pathway and who reviews decisions. Where a review is involved, the reviewer should be separate from the original decision-maker, and that separation should be visible to the person affected, not just true on paper.
Independence matters most where trust is most fragile. It is the difference between a review people accept even when they lose, and a review they assume was predetermined. Neutrality in a process does not guarantee fairness, so test fairness directly: did people understand the process, have the opportunity to be heard, and receive reasons for the outcome?
For financial services firms, ASIC's Regulatory Guide 271 on internal dispute resolution is a clear example of regulators requiring defined processes, timeframes and the identification of systemic issues. Even outside that sector it is a useful benchmark for how explicit a system should be.
Capability Is What Makes the Design Work
Systems fail when the people who must use them lack the skills to do so. Frontline staff and managers are the system in practice. If they cannot name an issue, ask the right questions, hold a structured conversation and document a next step, the best-designed pathways will sit unused.
This is why I treat dispute system design as an organisational capability question, not a policy-writing exercise. A policy tells people what should happen. Capability determines whether it does.
1. What capability means at each level
Frontline staff: Calm, clear and fair responses to emotionally charged contact; the ability to clarify an issue, understand what matters and decide whether it can be resolved on the spot.
Managers and team leaders: Early conversations, coaching, clear management decisions and knowing when a matter is beyond their remit.
Triage and complaints officers: Consistent application of criteria, fair process and clear documentation.
Senior leaders: Visible support for early resolution, and willingness to hear what the system reveals.
2. Training must be specific to the team
Training should never be abstract. It should give people language, structure and confidence they can use in the next difficult conversation they have at work. That is why the work is most effective when it is designed for a specific team's legislation, context and real conversations.
Two examples from my own practice show what this looks like. For a federal government department, I designed and delivered a five-day accredited mediation training program in-house, tailored to the department's statutory context, so tribunal-facing staff were trained together inside their own regulatory frame rather than attending generic external courses.
For an ombudsman office, I delivered in-house communication and early-resolution training built on the office's real case patterns. The complaints teams gained shared language and structure for their hardest conversations, without leaving the building.
Shared language matters more than it sounds. When a whole cohort learns the same structure at the same time, handovers improve, escalation decisions become more consistent and people stop improvising under pressure.
3. New pressures are testing existing systems
Capability also has to keep pace with the environment. Complaints teams are now seeing AI-generated submissions, which raise questions about fairness, evidence integrity and workload that most procedures never anticipated. I built a dedicated program, Navigating AI in Complaints and Dispute Resolution, because the demand for practical guidance was clear. It now runs both as public sessions and as tailored in-house delivery for regulators and complaints teams. A system that cannot absorb new contact patterns will quietly degrade.
Training on its own is not enough either. What helps is follow-up: practice in real matters, short refreshers and a way for managers to bring difficult cases back for discussion. Embedding is what separates a workshop from a change in how work is done.
Feedback Loops: Using Outcomes to Surface Systemic Issues
A feedback loop takes what each matter reveals and routes it to the people who can change the underlying cause. Without one, a system resolves cases but never improves. With one, it becomes an early-warning mechanism for workload, role clarity, communication and fairness problems across the organisation.
Most organisations close a matter and file it. The pattern across ten matters in the same team never reaches anyone with authority to act. Conflict is information, and a feedback loop is how that information is used.
What to track
Themes and sources: which teams, decisions, processes or policies recur in concerns.
Pathway use: which pathways are used, and which are never chosen.
Timeliness: time to first substantive response, not just time to closure.
Outcomes: whether matters were resolved at the earliest appropriate stage, and whether people returned.
Experience: how people felt about the process, including those who did not get the outcome they wanted.
The common mistake is tracking only closure time. A quickly closed matter may be a badly resolved one. Pair timeliness with outcome and experience measures.
Close the loop in the open
Create a simple register of systemic issues, assign an owner for each and report on progress at a set interval. De-identify case information appropriately and respect privacy and confidentiality obligations. When a theme leads to a change, tell the people who raised it. That single step builds more trust in the system than any communication campaign.
Build a scheduled review of the system itself. ISO 10002:2018, the international standard on complaints handling, frames complaint management around monitoring and continual improvement, and its guidance is a useful reference for what a review should cover. Whatever structure you choose, put it in the calendar. Systems that are not reviewed drift back towards habit.
Where External Support Fits
External support is most useful for three things: an independent diagnostic of how disputes currently flow, facilitated design workshops that bring the right people into one room, and in-house training that builds capability and embeds the system. It should supplement your internal ownership, not replace it.

A diagnostic brings distance. People inside a system often cannot see the workarounds they have normalised, and staff will sometimes say things to an outsider they would not say to a manager. A review of registers, data and pathways, combined with confidential conversations, gives a clear picture of where the design is working and where it is not.
A facilitated design workshop turns that picture into decisions: which pathways to build, what the triage criteria should be, where decision rights sit and how feedback will be reported. Getting complaints, HR, legal, operational leaders and frontline representatives in the same room avoids the common problem of a design that works for one function and fails for another.
In-house training then gives the design a chance to work. Programs are tailored to the team's legislation, context and real conversations, delivered in the room and followed up so the skills hold. Shiv Martin Consulting works with government and enterprise teams across Australia and New Zealand, including ombudsman offices, commissions, tribunals, regulators, departments and enterprise HR and complaints teams. Related services, such as conflict coaching, mediation and conciliation, can also sit inside the pathways you design.
If you are reviewing how disputes and complaints move through your organisation, a confidential conversation is the practical first step. You can book one with me to talk through your current system and how tailored consulting and in-house training could build your team's capability. The 40+ five star Google reviews our practice holds come largely from this kind of tailored, founder-led work.
What I Have Learned About the Stage Nobody Owns
The most consistent finding in my practice is that the early stage of a dispute is an ownership problem, not a policy problem. Organisations invest heavily in formal stages with named owners, timeframes and reporting. The first conversation, where resolution is cheapest, usually belongs to nobody.
That shapes how I advise on design. Before changing a procedure, I ask a blunter question: who is responsible for the first real conversation about this concern, and are they permitted, trained and expected to have it? If the answer is that it depends on who happened to receive it, the system is relying on luck.
I also hold a contrarian view on measurement. Most organisations report time to closure. I would argue time to first real conversation is the better leading indicator. Closure time rewards quick finishing, including quick dismissal. Time to first real conversation rewards the behaviour that prevents escalation: someone listening, clarifying and committing to a next step while the issue is still small.
The third lesson is that careful does not mean slow. Many leaders believe moving slowly is the safe, respectful way to handle conflict. In practice, delay often makes things worse, because people begin reacting to the silence.
A fast, clear, structured first response, even one that says what will and will not happen, is usually kinder than a thorough but late one. Clarity is kind. Good process creates safety, and safety is what lets people raise concerns while they are still small.
The practical implication for a head of complaints or HR director is modest to start. Pick one entry point, name an owner, define what that person may do without escalating, train them in a structure such as the Early Resolution Sequence, and track time to first real conversation for a quarter. You will learn more about your system from that single change than from a full policy rewrite.
If you want to build the capability that makes the early stage of your system work, you can read more about my training and workshops, which I run for complaints teams, frontline staff and managers. For people who handle concerns at first contact, I also offer Early Resolution Skills for Complaints and Case Management as a course.
FAQs
What is a dispute resolution system?
A dispute resolution system is the connected set of entry points, triage, pathways, escalation rules, people capability and feedback mechanisms through which an organisation handles concerns and disputes. It differs from a single process because it is designed end to end, with owners at every stage and a way to learn from outcomes.
What are the main components of an effective dispute system design framework?
The core components are accessible entry points, clear triage criteria, multiple pathways (coaching, facilitation, mediation, investigation and management decision), defined escalation and decision rights, trained staff and managers, and feedback loops with scheduled review. The weakest component usually determines how well the whole system performs.
How is an internal dispute resolution process different from mediation?
Mediation is one pathway within an internal dispute resolution process. The process or system includes everything around it: how a matter is raised, who decides if mediation is appropriate, what happens if it does not resolve the matter and how the organisation learns. Many matters are better served by coaching, facilitation, investigation or a management decision.
Does every dispute need mediation?
No. Not every matter needs the same process. Contested facts or possible misconduct may need investigation. A skills gap may need coaching. A shared team issue may need facilitation. Some matters need a clear management decision with reasons. Choosing the right process at the right time is the central design task.
How do I know if our dispute resolution system is working?
Look beyond closure times. Check whether matters are resolved at the earliest appropriate stage, how long it takes for someone to receive a first substantive response, whether all pathways are used, whether people return with the same issue and whether recurring themes lead to visible change. Staff and user experience should be part of the picture.
References
If you are reviewing how disputes flow through your organisation, send me an enquiry and we can talk through what would help.





.png)
