When conflict persists, adding another screen is tempting: a dashboard of indicators, a form for gathering opinions, an instant poll or an application that allocates shifts. The tool promises to turn disagreement into manageable data. Yet conflict rarely arises because an interface is missing. It usually appears because legitimate interests collide, experiences have not been recognised, material limits remain unnamed or responsibility is being displaced. In leadership and collective decisions, listening is not a courteous stage before the decision. It is the work through which people understand what is actually at stake, test reasons and assume an explainable choice.
A screen can show how many people prefer each option, but cannot by itself explain why they prefer it or what cost a minority would bear. It can record incidents without giving anyone authority to correct them. It can summarise hundreds of comments while erasing decisive nuances through classification. When management presents technology as the complete solution, it turns a political or organisational conflict into a question of use: a dissenter appears not to understand the system, to have completed the wrong field or to have overlooked the proper dashboard.
The alternative is not to reject digital tools. It is to assign them a proportionate function. They can preserve memory, widen channels, compare scenarios and make commitments visible. They cannot determine which harm is acceptable, which promise should prevail or who will answer for the consequences. Those questions require conversation, judgement and authority operating within limits. Process quality depends less on the sheen of an interface than on a genuine chance to influence the choice before it closes.
The map cannot replace the territory of conflict#
Every information system selects. It determines what becomes a field, which options are allowed, when the record closes and what remains outside the calculation. That selection may be reasonable, but it is never neutral. A dashboard measuring only time and cost will favour quick, cheap solutions even when they erode trust. One that counts only votes will confuse intensity with volume and overlook rights that should not depend on popularity. Before requesting more data, an organisation should therefore ask an earlier question: what reality is the system trying to represent, and who helped define it?
Every information system selects. It determines what becomes a field, which options are allowed, when the record closes and what remains outside the calculation.
The OECD links open government with transparency, integrity, accountability and stakeholder participation. That reference helps distinguish information from openness. Publishing a dashboard does not make a decision transparent when weighting criteria remain hidden, nobody knows who may alter the proposal or contributions arrive after the decisive moment. Useful transparency makes the reasoning reconstructable: what was known, which uncertainties persisted, which alternatives were rejected and for what reasons.
Listening begins by describing disagreement without reducing it to two camps. A meeting can open with a provisional formulation: one decision is pending, several needs are in tension and certain effects would be difficult to reverse. The description is then tested with affected people. If someone challenges the framing, the objection is not treated as a diversion from the agenda. It may reveal that the problem has been defined from the position of whoever controls resources or vocabulary.
Facts, interpretations, interests and proposals must also be separated. Facts can be checked; interpretations explain their significance; interests show what each party needs to protect; proposals are possible and revisable responses. A screen tends to combine all four layers in a box labelled comments. Skilled deliberation distinguishes them so that inaccurate data can be corrected without discrediting a legitimate need, and an unpopular proposal can be examined without assigning motives to the person who made it.
Listening is an institutional capacity, not a personal gesture#
Organisational listening requires time, roles and consequences. If people are invited to speak but nobody records commitments, the experience becomes release without influence. If opinions are collected and management never explains what changed, consultation breeds cynicism. If each manager listens by different standards, access to an answer depends on informal relationships. Individual goodwill matters, but only a stable structure ensures that reasons travel through a hierarchy to the place where power to act exists.
The International Labour Organization describes social dialogue as negotiation, consultation and exchange of information among governments, employers and workers. The lesson for other organisations is concrete: listening is not equivalent to sending a survey, because dialogue involves recognisable parties, negotiable matters and conditions for responding. When a decision affects the people performing work unevenly, their knowledge must enter before objectives, constraints and metrics are fixed.
A sound process identifies who convenes, facilitates, supplies evidence, decides and reviews. Several functions may be held by a small number of people, but they should not remain implicit. A facilitator protects speaking opportunities, clarifies points and notices silence; the role must not be used to steer the result. A decision maker listens and eventually chooses within the given mandate. A reviewer checks procedure, conflicts of interest and unexpected effects. Clarity prevents a situation in which everyone participates while nobody answers.
Participation should be designed in proportion to risk. A small and reversible choice may be settled through a short conversation and a trial. A measure that changes employment, access to services or personal-data processing demands more time, representation and review. Applying the same form to every question produces an appearance of consistency while ignoring seriousness. Deliberative intensity should rise with irreversibility, unequal impact and uncertainty.
Case one: the productivity dashboard that conceals care#
A home-care provider introduces a dashboard to reduce delays. Each visit appears in green or red according to its closing time, and teams receive weekly rankings. Within weeks, reported punctuality improves. Complaints about rushed conversations and unrecorded tasks also increase. Care workers explain that some visits require calming a disoriented person, waiting for medication or coordinating with relatives. The system reads those variations as individual inefficiency.
Management responds by adding a screen for explaining delays. The form contains a closed list, requires several fields at the end of each shift and does not amend the ranking until a supervisor reviews it. This new layer generates more data and more work, but does not settle the disagreement about what good care means. Workers learn to choose the least disputed category; managers receive tidy reports; service users continue to feel hurried. The conflict has been transferred from planning into the language of recording.
A way forward begins with small sessions where professionals, service users, coordinators and health-and-safety staff describe concrete situations. The group agrees to distinguish avoidable delay, clinical or relational need, and planning failure. It reserves buffers between certain visits, limits individual rankings and combines punctuality with continuity, completed tasks and qualitative assessment. The change is tested for six weeks in one area, and the indicators capable of producing employment consequences are disclosed.
The final decision does not satisfy every preference. The budget prevents every visit from being lengthened, and schedules still need control. Reasons are nevertheless explicit, exceptions enter the design, and a joint committee reviews effects. The screen stops pretending to define quality and starts showing one part of it. Leadership does not abandon responsibility behind the algorithm; it explains the balance selected, the harms it will watch and the point at which the decision will be reopened.
Case two: a neighbourhood consultation turned into a contest of clicks#
A municipality must choose whether to calm traffic on an avenue, retain parking spaces or expand room for café terraces. It launches a platform with three images and allows voting for fifteen days. The option with most support retains parking. Shopkeepers and residents welcome the clarity of the result, while disability groups and families from the local school say the process omits dangerous journeys and loading times. The authority proposes an additional map on which problems can be marked.
The map improves spatial evidence but preserves the original flaw: alternatives were defined before listening, and every click has equal weight without examining who bears risk. Moreover, many people who cross the avenue do not live in the district and never received an invitation. Treating conflict as a competition among preferences hides accessibility and safety duties. A majority may guide discretionary uses, but cannot turn a basic right into a subordinate option.
The Council of Europe includes participation, responsiveness, efficiency, openness and accountability among its principles of good governance. Applied here, those principles support reopening the frame. The team holds accessible street walks, observes school arrival times, listens to delivery workers and publishes technical constraints. It then creates two feasible alternatives that guarantee safe routes and submits remaining differences to deliberation: the number of spaces, construction phases and loading periods.
The resolution reserves a timed loading area, reduces parking, widens pavements and trials a terrace rule for three months. It is not announced as the neighbourhood’s mathematical will. It is presented as a reasoned public decision combining evidence, rights and preferences, with monitoring measures. Dissenters can identify the applied criteria and request review if use data contradicts forecasts. The process gains legitimacy not because disagreement vanishes, but because disagreement is no longer hidden behind a counter.
Case three: the internal portal that multiplies requests and silence#
A university merges two departments and must allocate laboratories, teaching and administrative support. To avoid endless meetings, it opens a portal where every unit enters needs and attaches documents. The committee receives hundreds of records and an automatically generated priority matrix. Groups with administrative experience submit detailed cases; a young team with short contracts and a heavy teaching load provides less information. The matrix interprets the difference as lower need.
When protests arise, an appeals inbox is added. The committee receives even more text but keeps its original timetable and criteria. Nobody speaks with teams to check dependencies: a shared laboratory requires particular ventilation, one course concentrates practical work in six weeks, and several duties depend on a single technician. Documentary abundance creates a feeling of rigour while the relationships among data remain unseen.
The institution suspends irreversible allocations and convenes tables with balanced representation. Every group receives equal support for preparing evidence. A shared inventory of constraints is built, conflicts of interest are declared, and minimum needs are distinguished from negotiable preferences. The committee publishes a provisional proposal, takes oral and written questions and corrects errors before deciding. The portal retains documents and history, but no longer acts as a substitute for encounter.
The UNESCO programme on access to knowledge recalls the importance of conditions that enable people to access, produce and share information. In this situation, access means more than possessing a password. It includes understanding categories, having time to supply evidence and reaching a channel capable of resolving questions. The portal’s formal equality was insufficient because teams began with different resources and positions.
The resulting allocation retains painful limits: some spaces are shared and certain preferences are postponed. Common criteria, transition support and a six-month review are nevertheless established. Disagreement leaves a useful memory instead of a collection of defensive forms. The university can explain both the allocation and the route taken, while teams know what information would justify a future change.
Turning information into deliberation#
The first practical step is to write a decidable question. People are not asked vaguely what they think about a project, but which alternative can meet a purpose within known limits. The invitation explains which aspects are open, which are imposed by law or budget and who will make the decision. Such precision reduces false expectations and helps contributions concentrate on relevant reasons.
The second step is to compose the table. A universal invitation is not enough when time, language, hierarchy or digital access produces actual exclusion. The organisation identifies affected groups, frontline knowledge and voices that will carry costs. It offers written and oral formats, different times, accessibility and advance documents in plain language. The aim is not to give every identity a fixed spokesperson, but to prevent ease of participation from being mistaken for representativeness.
The third step is to return what has been heard. A provisional synthesis attributes arguments without exposing personal data, identifies agreements, tensions and gaps in evidence, and allows misunderstandings to be corrected. The authority then decides and publishes a response connecting reasons to concrete changes. The formula is not “this was taken into account”, but “this contribution changed this element; that one was not incorporated because of this constraint”. Traceability turns participation into verifiable influence.
Tools can support all three steps. A form gathers comparable evidence; a video call brings in people at a distance; a dashboard tracks commitments; an archive preserves versions. Each tool needs an alternative, a responsible person and a coherent closing date. Participation should not disappear when a channel fails. If automated classification summarises comments, a person must review lost meaning and explain the system’s limits.
Deciding, answering and listening again#
The European Ombudsman highlights public-service principles including objectivity, respect, transparency and accountability. These principles prevent listening from being presented as a favour granted by management. Whoever exercises power must be able to justify the procedure, the information used and the treatment of comparable arguments. A review route must also exist without depending on private persistence or knowing the right person.
An explainable decision includes its purpose, alternatives, evidence, criteria, uncertainties, responsible authority, review date and means of repairing unexpected effects. It need not become an endless file. A short document can fulfil the purpose if it honestly distinguishes what is known from what is assumed. The record protects participants and decision makers because it reduces the temptation to rewrite reasons retrospectively.
Monitoring should measure effects, not merely activity. Numbers of surveys, portal visits or meetings reveal little about the distribution of benefits and burdens. It is worth observing who withdrew, which groups changed position after receiving information, which commitments were fulfilled and which harms emerged. Quantitative data should be interpreted alongside testimony and direct observation. When sources disagree, the response is not to select the most convenient indicator, but to investigate which part of reality each source captures.
Finally, listening after a decision does not mean reopening everything permanently. It means setting review conditions: a date, a threshold of harm, new evidence or a change of context. It also means communicating which elements will remain stable so people can act predictably. Democratic leadership does not refuse closure. It closes provisionally where uncertainty exists and firmly where obligations are clear.
Another screen may ease a specific difficulty, but becomes evasion when it prevents an organisation from naming conflict, gathering parties or assuming a choice. A mature organisation first asks which relationship is failing, which power remains hidden and which reasons do not reach the deciding space. It then chooses the smallest tool that helps people understand and remember. The result will not be automatic unanimity. It will be something more valuable: a decision that acknowledges its costs, can be challenged with evidence and has someone prepared to answer for it.
