How do I build trust across organisational boundaries?

Trust across organisational boundaries is not the same thing as trust within a team. It is built differently, damaged differently, and sustained differently. Most lateral leaders apply within-team trust-building behaviour to cross-boundary relationships and are puzzled when it does not work as well.

There is also an economic argument worth making before anything else. Low trust environments are measurably slower and more expensive than high trust ones. Every cross-boundary interaction that lacks trust generates a tax: the time spent on verification, the energy spent on managing defensiveness, the rework that happens when people do not feel safe enough to raise a concern early. In a matrix, where every significant outcome depends on people who do not report to you, that tax compounds quickly. The lateral leader who builds genuine trust across boundaries does not simply have better relationships. They get more done, faster, with less friction, because the trust dividend is real and it accumulates over time.

I was working with a senior IT transformation leader who had just been appointed to lead a three-year digital programme across a pan-European manufacturing organisation. She had no formal authority over any of the five functional heads whose cooperation the programme depended on. Two of them had been through a previous technology programme that had gone badly, a botched ERP implementation that had overrun by eighteen months and left a trail of damaged relationships and unrealistic expectations. She was not starting from a blank slate. She was starting from a deficit.

Her instinct was to move quickly. She prepared thorough briefings, ran detailed stakeholder sessions, and set out a clear programme plan within her first month. At the end of month two, she asked me why three of the five functional heads still felt, as she put it, like they were watching her rather than working with her.

The answer was that she had been building the wrong kind of trust for the wrong stage of the relationship. She had been proving her ability in detail before she had established her integrity in principle, and she had been demonstrating what she knew before she had shown she genuinely cared about what they needed. In a context where two of her key stakeholders had already been let down by someone who had arrived with the same confident competence, competence alone was not going to move them. Two of them were not simply cautious about her. They felt, however unfairly, that a commitment had been broken by her programme’s predecessor, and that feeling was shaping every interaction regardless of what she did.

Trust across boundaries is not one thing

Consider two matrix leaders, each needing to build trust with a function they have no formal authority over.

The first treats trust as a single, undifferentiated thing to be accumulated over time. She is professional, consistent, and delivers what she promises. Six months in, she has a solid reputation within her own programme team but has made limited progress with the functional heads she needs most. She has not distinguished between the different things trust actually consists of and so has been building some of them well and neglecting others entirely.

The second understands that what she is being asked to build in each relationship is a specific combination of things, and that the combination differs across relationships and changes at different stages. She uses a simple diagnostic before each significant relationship investment: what, specifically, is this person uncertain about? Is it whether she can deliver? Whether she cares about their interests? Whether she will behave consistently? Whether she will share information honestly? The diagnosis changes what she does.

The research on organisational trust offers a precise framework for this. Trustworthiness, in any significant relationship, is a function of credibility, the confidence that what you say is grounded and honest; reliability, the confidence that you will do what you say; intimacy, the sense that it is safe to be honest with you about difficult things; and self-orientation, the degree to which you appear to be acting in your own interests rather than the other person’s. The last of these is the denominator: it does not simply reduce the trust that the other elements are building, it actively destroys it. A moderately credible, moderately reliable leader who genuinely orients towards the other person’s interests will generate more trust, faster, than an exceptionally capable leader who is visibly focused on what they need.

Underneath these elements is a further distinction worth making, particularly for leaders managing relationships where trust has previously been damaged. There are at least three types of trust operating simultaneously in any cross-boundary relationship: trust in your competence and your ability to deliver, trust in your communications, specifically whether you share information honestly, admit mistakes readily, and maintain confidences reliably, and trust based on your agreements, whether you keep what you commit to, whether you behave consistently whether or not you are being observed. Each type is built by different behaviours, is damaged by different events, and is repaired by different actions. When a relationship is not developing as it should, the useful diagnostic question is not “how do I build more trust” but “which specific type of trust is most at risk in this relationship, and what is the specific behaviour that would address it.”

There is also a timing issue that most lateral leaders underestimate. Cross-functional programmes have timelines. The trust that a lateral leader needs in month ten, when a difficult trade-off needs to be navigated, has to have been built in months one through nine. Trust that is only sought at the point of need is already too late. And extending trust first, being willing to trust before you have full evidence that it will be reciprocated, is one of the most reliable accelerators available in a new cross-boundary relationship. It signals confidence rather than caution, and it invites reciprocity in a way that a measured, wait-and-see approach does not.

This article explores how to build trust across organisational boundaries deliberately and sustainably, drawing on the behavioural science of trust formation, credibility, and cross-boundary relationship development. It is part of a longer series on matrix leadership. The foundation article, How do successful leaders create commitment in matrix organisations?, sets out the full framework. The series also covers what a matrix organisation is and why companies adopt one, how to build influence without authority, how to build a powerful internal network, how to convene people who don’t report to you, how to create accountability without authority, how decisions get made in a matrix organisation, how to manage competing priorities in a matrix, and how to resolve conflict between departments.

Why trust is so hard to build across boundaries in a matrix

Starting from the wrong kind of trust: In a new cross-boundary relationship, most leaders lead with their competence: demonstrating what they know, what they have done, and what they plan to deliver. This builds competence-based trust, which matters, but it is not the element most cross-boundary relationships are actually lacking. The functional head who has been involved in a failed programme is not primarily doubting your ability. They are doubting your intentions and your consistency. Leading with competence in that context addresses the wrong problem.

Treating trust as a destination rather than a dynamic: Trust is not a state that is reached and then maintained. It is a pattern of interactions, each of which confirms or undermines the signals the other person is reading. A lateral leader who builds trust steadily and then stops actively investing in it, because the relationship feels secure, will find that the trust quietly erodes, not through any dramatic event but through the accumulated signal that the relationship is now less important to them than it was.

High self-orientation: This is the most common and least noticed trust problem in matrix leadership. A lateral leader who is visibly focused on what they need from a stakeholder, who arrives with a well-prepared pitch about their programme’s requirements, who follows up consistently about the things that matter to their own delivery, is sending a clear signal about their orientation whether they intend to or not. Self-orientation is the denominator in any trust relationship and its effect is not linear. Even small increases in perceived self-orientation can erase large amounts of trust built through credibility and reliability.

Confusing reputation with relationship: A strong professional reputation is a useful starting point in a new cross-boundary relationship, because it provides an initial basis for trust before any personal history has been built. But reputation is borrowed trust. It gives you a starting position. What you do with that starting position, in the first interactions, is what determines whether the relationship develops or stalls.

Building trust with people who chose the relationship while neglecting those who did not:  Cross-boundary trust tends to be built selectively: with the stakeholders who are supportive, engaged, and easy to work with. The stakeholders who are sceptical, distant, or carrying the weight of a previous disappointment are precisely the ones who need the most deliberate trust-building investment, and they most often receive the least.

Not distinguishing between trust that is absent and trust that has been broken:  When a relationship is not progressing, the usual diagnosis is that not enough trust has been built yet. But in a matrix, the more common and more serious problem is trust that was damaged before the current leader arrived, either by a previous programme, a previous relationship, or a broken organisational commitment that has nothing to do with the current leader personally. Absent trust and broken trust require fundamentally different approaches. Absent trust is built by consistent positive signals over time. Broken trust requires acknowledgement of what happened, specific repair actions, and a demonstrated change in the conditions that led to the damage. Treating broken trust as though it were simply absent trust, by doing more of the same positive things, is one of the most common and most frustrating cross-boundary trust failures.

Six practices for building trust across organisational boundaries

Diagnose which type of trust is at stake before investing in the wrong one

The most efficient trust-building investment is the one that addresses the specific element of trust that is actually missing. But most lateral leaders do not make that diagnosis. They invest in general trust-building effort across a relationship and wonder why it is not moving the needle in the specific direction they need.

The diagnosis begins with a simple question: what, specifically, is this person uncertain about? Are they uncertain about whether you can deliver what you have said? That is a competence trust problem, and the solution is relevant evidence of past delivery combined with a clear and honest account of how you are approaching the current challenge. Are they uncertain about whether you will share information with them honestly, tell them when something is going wrong before it becomes a crisis, and maintain confidentiality about what they share with you? That is a communication trust problem, and it is addressed by the specific behaviours of proactive disclosure, honest admission of difficulties, and demonstrable discretion. Are they uncertain about whether you will keep your agreements, behave consistently whether or not you are being watched, and do what you say regardless of convenience? That is a contractual trust problem, addressed by the consistency and specificity of your commitments and the rigour with which you follow through on all of them, not just the ones that are visible.

Research on how integrity develops in relationships is particularly relevant here: it is most important early on, before any personal history has been built, because in the absence of shared experience it is the element that determines whether the other person will extend enough initial trust to allow the relationship to develop. In a matrix, where relationships often have to function quickly, a very early signal of inconsistency between what a leader says and what they do can close down a relationship before it has really opened.

A programme manager in a global logistics company had spent six months building a cross-boundary relationship with a regional operations director through consistent, professional delivery. The relationship was functional but had not become collaborative. When she worked through the three types carefully, it became clear that the operations director was not questioning her competence or her consistency. He was questioning her orientation: whether she was there to serve the programme or to serve the business. Every interaction had been about what the programme needed from operations. He had no evidence that she was equally interested in what operations needed from the programme. She had been building competence trust and contractual trust for six months while the real deficit was in how he read her self-orientation.

The IT transformation leader worked through the same analysis across her five functional heads and found that the pattern was different for each one. Two of the functional heads who had been through the failed ERP were primarily uncertain about her communication trust, whether she would tell them when something was going wrong before they heard it elsewhere. One was primarily doubting her orientation, whether the programme was designed around the business’s needs or around the IT function’s capability agenda. Two others were broadly confident in her competence but were watching to see whether the early positive signals would be sustained or would turn out to be a honeymoon period.

Questions you can use to diagnose which type of trust is at stake:

• When this person expresses reservation about working with you, what is the reservation actually about: your capability, your communications, your consistency, or your orientation towards their interests?
• Is there any gap between what you said you would do and what you have actually delivered that this person has noticed but not named?
• Does this person have clear evidence that you share information with them honestly and early, including when the information is difficult?
• How does this person describe you to others: in terms of your competence, your reliability, or your intentions?
• If you asked this person directly what would make them more willing to work closely with you, what do you think they would say?

Practical actions:

• At the start of any significant cross-boundary relationship, spend time understanding what the other person’s previous experiences with similar partnerships or programmes have been. What they tell you is a map of which type of trust they are not yet ready to extend and why.
• After each significant interaction, ask yourself which element of trust that interaction was most likely to have affected, positively or negatively, and whether the effect was the one you intended.
• When a cross-boundary relationship is not progressing despite sustained effort, stop and diagnose which type of trust is actually missing rather than simply intensifying the effort in the same direction.
• Ask a trusted colleague who knows the stakeholder what they understand the relationship concern to be. The diagnosis from outside the relationship is often more accurate than the one from inside it.

Use early trust wisely and extend it first

In a new cross-boundary relationship, before any personal history has been built, there is usually a brief period in which the other person extends provisional trust based not on what they know about you specifically but on what they expect from someone in your role, with your background, from your organisation. This is borrowed trust. It has a short shelf life, and most lateral leaders waste it.

Research on trust formation in temporary and cross-functional systems is clear about this: initial category-based trust, the kind based on professional role and reputation, is enough to get a relationship started but is fragile. It erodes quickly with any signal of inconsistency. Most lateral leaders spend the early period of a relationship demonstrating what they know. They brief stakeholders on the programme, explain the approach, show the plan, and demonstrate their command of the detail. This is not wasted effort, but it is a relatively inefficient use of the early trust that is available. The most effective use of that early period is to listen more than to speak: to understand what the stakeholder is working on, what they are trying to accomplish, where they feel the organisational pressure most, and what they need from the programme that they have not yet been able to ask for.

There is a second and more active move available here, one that most lateral leaders are reluctant to make: extending trust first. Rather than waiting for a relationship to develop to the point where trust feels warranted, offering genuine trust early, delegating something real, sharing something candid, treating someone as an ally before they have fully demonstrated they will act as one, is one of the most reliable trust accelerators available. It is not naive. It is strategic. Trust offered before it has been fully earned invites reciprocity in a way that a carefully measured, wait-and-see approach does not, and it sends a signal about the kind of relationship you are prepared to have that a defensive posture never can.

An organisational development lead in a global financial services firm described starting a new cross-functional role by spending her first eight weeks doing nothing except listening conversations with each of the functional leaders she would need to work with, asking specifically about what was currently getting in their way and what they needed that they were not getting. She did not mention her own agenda in any of those conversations. At the end of eight weeks, she had three pieces of work she could do immediately that would make a material difference to people she needed on her side. She did those pieces of work first, before advancing her own agenda. The trust she built in that period, by using her early access to listen and to serve rather than to pitch, was substantially more durable than anything she had built in previous roles through the standard stakeholder management approach.

The IT transformation leader changed her approach in month three. She stopped leading with the programme plan and started every significant meeting with a version of the same question: what is making your work harder right now and what would make it easier? Within six weeks, two of the five functional heads had come to her with a problem they had not taken elsewhere. That shift, from programme representative to genuine ally, would not have happened through any amount of programme briefing.

Questions you can use to make the most of early trust in a new cross-boundary relationship:

• In your first five to ten interactions with a new cross-boundary stakeholder, what proportion of that time have you spent listening rather than presenting?
• Do you know more about what this person is trying to accomplish and what is making it difficult than you know about their formal role and their position in the organisation?
• Is there anything you could do in the next two weeks that would make a material difference to this person’s work and that they are not expecting from you?
• Have you been willing to extend trust first, to treat this person as a genuine ally before they have fully demonstrated they will act as one?
• What signal, if any, have you sent in early interactions that might have begun to erode the initial trust rather than develop it?

Practical actions:

• In the first month of any significant cross-boundary relationship, prioritise listening conversations over briefing sessions. Ask about challenges, pressures, and unmet needs before presenting your own agenda.
• Identify at least one thing you can do early in the relationship that serves the other person’s interests at no cost to your own, and do it before you have asked for anything in return.
• Find one deliberate opportunity to extend trust before it has been fully earned: share something candid, delegate something real, or treat the other person as a partner in a decision before the relationship has fully warranted it. Notice what it generates.
• At the three-month point in any new significant cross-boundary relationship, review whether the provisional trust you started with has developed or is being maintained at roughly the same level. Stagnation at the initial level is a warning sign.

Reduce your self-orientation before everything else

The single most reliable predictor of whether a cross-boundary relationship develops beyond professional courtesy into genuine working partnership is not how competent the lateral leader is or how consistently they deliver. It is how self-oriented they appear to the other person.

Self-orientation is the denominator in the trust relationship: it does not simply reduce the trust that credibility and reliability are building, it actively destroys it. A matrix leader with moderate competence and a genuine orientation towards the other person’s interests will build more trust, faster, than an exceptional leader who is visibly focused on what they need from the relationship. The challenge is that self-orientation is often invisible to the person who has it. Programme pressure creates a natural pull towards focusing on what the programme needs, and a leader who is entirely well-intentioned can come across as extractive simply because every significant contact they initiate is about something they require.

In a cross-boundary context, reducing self-orientation is not simply a matter of caring more. It is a set of specific, observable behaviours. Going into a stakeholder meeting with a genuine question about what they need before presenting what you need. When a trade-off arises between what the programme requires and what the other function needs, being seen to advocate for the other function’s position rather than defaulting to the programme’s. Following up on the things that matter to them with the same urgency as the things that matter to you. These are not gestures. They are signals that accumulate into a clear picture of whose interests you are actually serving.

A commercial director I worked with described a cross-boundary relationship with a supply chain function that had been superficially cooperative for two years without ever becoming genuinely collaborative. When he looked at the pattern of his interactions, almost every significant contact he had initiated had been about something the commercial function needed from supply chain. He had been reliable, professional, and responsive to their requests. He had never, unprompted, asked what supply chain needed from commercial or done anything that served their interests at a visible cost to his own. The relationship reflected that exactly: supply chain cooperated but did not partner, because they had no evidence that the relationship would ever work in their direction.

The IT transformation leader recognised this pattern clearly in her own approach. Every meeting she had prepared for had been structured around the programme’s needs. Her follow-up communications had been about programme deadlines, programme requirements, and programme decisions. When she began to restructure her interactions, asking about the functional heads’ own pressures and following up on those with the same rigour as she followed up on her own, the change in several of the relationships was noticeable within a month.

Questions you can use to assess and reduce your own self-orientation in cross-boundary relationships:

• In your last five interactions with this stakeholder, what proportion of the agenda was driven by what you needed versus what they needed?
• Do you follow up on the things that matter to this person with the same speed and rigour as the things that matter to your programme?
• When a trade-off has arisen between your programme’s interests and this function’s interests, have you ever been seen to advocate for their position rather than your own?
• If this person were asked to describe what you are most interested in, what would they say?
• What have you done in the last month that served this person’s interests at a visible cost to your own?

Practical actions:

• Before every significant cross-boundary interaction, write down one question you will ask about what the other person needs before you raise anything about what you need.
• When following up after meetings, include at least one item that is about their work and their priorities, not just the programme commitments and decisions you need to track.
• Once a quarter, ask yourself whether each key cross-boundary stakeholder has clear evidence that you care about their success independently of whether it serves your programme. If the answer is uncertain, it is probably no.
• Find one occasion per month where you can be seen to advocate for a cross-boundary stakeholder’s position in a forum where it costs you something to do so. The visibility of this matters as much as the action itself.

Build consistency across all contexts, not just the visible ones

In a hierarchy, trust is primarily built through direct, bilateral experience. In a cross-boundary matrix context, this is only half the picture. A significant proportion of the trust that exists in a cross-boundary relationship is based on what each person hears about the other through the network. A lateral leader who behaves impeccably in a relationship but inconsistently in the broader network is building trust in the room and destroying it outside it.

Contractual trust, the kind built through consistent behaviour and kept agreements, operates across all contexts simultaneously. It is not enough to keep your commitments to the people you are most trying to build trust with if you are visibly inconsistent in your commitments to others. The network observes both. A leader who applies different levels of rigour to commitments depending on who is watching, who says different things in different rooms, or who frames the programme’s relationship with one function differently depending on which function they are talking to, will find that reputation information about that inconsistency travels faster than any positive signal they are building directly.

This is particularly important at the start of a programme, when the lateral leader’s reputation is being actively formed across the network. The early impressions that form in each function are quickly shared across functions through the informal channels that any established group of colleagues uses. A lateral leader who is seen to treat one function differently from another will find that their reputation calcifies into something they did not intend and that is very hard to change later.

A strategy director described a cross-boundary relationship that had never quite recovered from a single incident in her first month. She had spoken critically about one function’s readiness in a meeting with another function’s leadership, thinking the conversation was informal and would not travel. It did. The function she had spoken about heard a version of it within a week. The direct relationship continued, professionally, but it never became the collaborative partnership she needed.

The IT transformation leader made a version of this mistake in month two. She had given a programme update to the group’s technology steering committee that was noticeably more optimistic than the update she had given to two of the functional heads the week before. One of them was also on the steering committee and noticed the difference. She addressed it directly and immediately: she corrected the steering committee record, and she spoke to the functional head directly about what she had done and why. The repair was not cost-free but it was faster than it would have been if she had left the inconsistency unaddressed.

Questions you can use to audit your consistency across contexts:

• Is the story you tell about this programme and these relationships the same in every room, or does it change depending on the audience?
• If the five functional heads in your matrix compared notes on what you have said to each of them, would the accounts be consistent?
• Are there commitments you have made to people who are less senior or less visible that you are tracking with the same rigour as commitments to more senior stakeholders?
• Is there anything you have said or done in a context you considered informal that, if it reached the person you are most trying to build trust with, would damage the relationship?
• What is your reputation in the network more broadly, and is it consistent with how you are in the direct relationships you are most invested in?

Practical actions:

• Treat every interaction in the network as potentially visible to the person you are most trying to build trust with. Behave as though the conversation will be reported, because in most established organisations it will be.
• Apply the same standard of rigour and follow through to commitments made to operational and less senior colleagues as to those made to senior stakeholders. The network observes both.
• If you have said or done something inconsistent across contexts, address it directly and quickly before it surfaces through the network. A self-corrected inconsistency damages trust far less than one that is discovered.
• At regular intervals, ask a trusted colleague who has visibility of your behaviour across different contexts whether your reputation in the network is consistent with how you behave in the relationships you value most.

Build trust at the speed of vulnerability, not only the speed of competence

Demonstrating competence is the most common and the least efficient trust-building strategy available to a lateral leader. It is necessary: a cross-boundary stakeholder who doubts your ability will not give you the cooperation you need. But it produces a specific and limited kind of trust, the kind that says this person can do the job. What it does not produce is the kind of trust that says I can tell this person something difficult and it will be handled well, or this person will tell me when something is going wrong before it becomes a crisis.

That deeper trust, which makes cross-boundary relationships genuinely resilient rather than just professionally functional, develops through communication trust: the confidence that you will share information honestly and early, that you will admit what you do not know before you have had a chance to find out, and that you will name a difficulty directly rather than managing how it lands. These are not signs of weakness in a cross-boundary context. They are the primary signals of integrity that most lateral leaders underuse, because they have been trained to project competence and are uncertain what signal admitting uncertainty will send.

The matrix leader who creates the conditions for honest exchange, typically by modelling it first, builds trust at a qualitatively different level from the one who simply demonstrates how good they are at their job. And the research on psychological safety is relevant here in a specific way: a cross-boundary relationship in which both parties feel safe to name a concern before it becomes a problem is a fundamentally more useful relationship, not just a warmer one.

A finance transformation lead described the moment that most changed a difficult cross-boundary relationship with an operations director. She was presenting a financial modelling approach in a senior forum and was asked a question about a set of operational assumptions she had not fully worked through. Rather than deflecting, she said directly that she did not know the answer to that question, that it was a gap in her current thinking, and that she would work through it with the operations director and come back with a properly considered response. The operations director, who had been watching the session with some scepticism, approached her afterwards and told her it was the first time in any cross-functional forum he had seen a finance lead admit to not knowing something in public. The relationship changed significantly from that point.

The IT transformation leader built her most important relationship in the programme through a similar move. One of the functional heads most affected by the previous failed ERP asked her directly, early in the programme, whether the current programme had the same risk profile as the previous one. Rather than reassuring him with a confident comparison, she said she did not yet know enough about what had gone wrong to answer that question honestly, and that she wanted to understand it properly before giving him an answer she could not stand behind. She spent two weeks working through the previous programme’s failure points, came back with a direct assessment that included three specific risks she could see in the current programme, and told him what she intended to do about each one. He became one of her strongest advocates in the steering group.

Questions you can use to build communication trust through vulnerability:

• When was the last time you admitted to a cross-boundary stakeholder that you did not know the answer to something, and what was their response?
• Is there a difficulty in the current programme that you are managing internally rather than naming honestly with the stakeholder who needs to know about it?
• What would you need to believe about this relationship to feel safe enough to name a concern directly and early rather than after you have resolved it?
• Are there questions this stakeholder is asking that you are deflecting because you do not have a good answer, rather than naming that directly?
• What would change in this relationship if you were consistently the first person to name a problem rather than the last?

Practical actions:

• Identify one thing in your current programme that you do not fully know the answer to, and name that directly with the relevant cross-boundary stakeholder before you have resolved it. Notice the response.
• When a cross-boundary stakeholder asks a difficult question in a forum, resist the instinct to give the most defensible answer available. If you do not know, say so, and be specific about how and when you will come back with a properly considered response.
• When something in the programme is not going as planned, tell the relevant functional heads before they hear it from another source. The timing of disclosure is one of the strongest signals of communication trust available to a lateral leader.
• Build a habit of ending significant interactions with a cross-boundary stakeholder by asking what concern has not yet been named in the conversation. The question creates space for the kind of exchange that builds deep trust.

Repair broken trust specifically, not generally

The trust that exists in a cross-boundary relationship before it has been tested is fundamentally different from the trust that exists after it has been tested and survived. Most lateral leaders invest heavily in building trust during smooth periods and find that a significant disruption resets much of what has been built. This is not inevitable. It is a consequence of how the disruption is handled rather than the disruption itself.

What makes this more complex in a matrix is that a lateral leader frequently inherits broken trust, relationships where a previous programme, a previous leader, or a previous organisational decision has already caused a genuine breach. This is different from absent trust, and it requires a different response. Attempting to build trust normally in a relationship where a breach has occurred and has not been acknowledged is one of the most common reasons cross-boundary relationships stall despite consistent positive effort. The unacknowledged breach sits in the room at every subsequent interaction, shaping how everything is interpreted, regardless of how well the new leader performs.

The research on trust betrayal and healing is specific about what repair actually requires: it is not simply a matter of demonstrating trustworthy behaviour over time, though that is part of it. Effective trust repair requires acknowledging what happened and its impact, taking clear ownership of the contribution to the breach (including inherited contributions that are not the current leader’s personal fault but are nonetheless the current leader’s responsibility to address), taking specific actions to repair the damage, and demonstrating a genuine change in the conditions that produced the breach. General positive behaviour in a relationship where a specific breach has occurred is insufficient. The specificity of the repair has to match the specificity of the damage.

There is also an important sequence: acknowledgement before action. A lateral leader who moves quickly to demonstrate trustworthiness without first acknowledging what was broken will find that the positive signals are filtered through the unaddressed breach and carry less weight than they should. In some cases, positive performance in an unacknowledged breach context can actually deepen scepticism, because it looks like an attempt to move past something that deserved to be addressed directly.

A project director leading an infrastructure rollout in a global retail business described a delivery failure that she had been certain would end two of her most important cross-boundary relationships. A systems migration had gone wrong in a way that disrupted operations in two regions for forty-eight hours. She had been aware of the risk the day before it materialised and had not escalated it. When the disruption happened, her first call was to both regional operations directors, before she had a resolution plan, to tell them exactly what had happened, what she knew, what she did not know, and what she was doing. She was in both cases the first person to reach them with a direct account rather than a managed one. Both relationships were stronger six weeks later than they had been before the incident, because both directors had seen exactly how she behaved when something went seriously wrong.

The IT transformation leader had to address inherited broken trust directly and early. In her third month, she convened a conversation with the two functional heads most affected by the previous ERP failure, not to discuss the current programme but to discuss what had happened in the previous one and what had made it damaging to the relationship between their functions and the IT function more broadly. She did not defend the previous programme. She acknowledged the breach, named its impact on the relationship, and asked specifically what would need to be different in the current programme for the working relationship to feel genuinely different rather than provisionally cautious. The conversation was uncomfortable. It was also the most trust-generative interaction she had in her first six months.

Questions you can use when addressing broken trust specifically:

• Is there a breach in this relationship, inherited or recent, that has not been directly acknowledged and that is shaping how every subsequent interaction is interpreted?
• Have you named the impact of the breach on the other person, not just the fact that something went wrong?
• Is the repair you are making specific enough to match the damage that was done, or are you hoping that general positive behaviour will be sufficient?
• What specific change in the conditions that produced the breach have you been able to demonstrate, as opposed to simply demonstrating that you personally would have done things differently?
• Is there anyone else who needs to be part of the acknowledgement and repair, because the breach was organisational rather than just interpersonal?

Practical actions:

• When you inherit a relationship where a previous breach has occurred, address it directly and early rather than hoping that good performance will gradually overwrite the history. Name what happened, acknowledge the impact, and ask what would need to be different.
• When a trust-damaging event occurs that you are responsible for, contact the affected stakeholder directly and early, before you have a full resolution plan but after you have enough clarity to describe what happened and what you are doing.
• When owning a failure or a difficulty, be specific about what you did or did not do that contributed to it. General acknowledgements of responsibility are less trust-building than specific ones.
• After any significant breach and repair, schedule a direct conversation four to six weeks later to review what has changed, not just to close the episode but to demonstrate that the conditions that led to it have been genuinely addressed rather than simply survived.

Wrapping up

Building trust across organisational boundaries in a matrix is not the same as building trust within a team, and treating it as though it were produces exactly the frustrating experience of working consistently hard at a relationship and feeling like it is not moving. The lateral leader who diagnoses which type of trust is actually at stake, who uses early relationship access to listen and extend trust first rather than to brief and wait, who genuinely reduces their own self-orientation rather than simply managing its appearance, who behaves consistently across all contexts and not just the visible ones, who builds deep communication trust through honesty and vulnerability, and who repairs broken trust specifically rather than hoping that positive performance will gradually overwrite the damage, is building something qualitatively more durable than professional goodwill.

That durability matters for a reason that goes beyond the quality of the relationships themselves. Low trust cross-boundary relationships generate a real and measurable cost in verification time, defensive energy, rework, and missed early warnings. High trust cross-boundary relationships are faster, lighter, and more resilient. In a matrix, where every significant outcome depends on people who do not report to you, the trust dividend is not a soft benefit. It is the primary determinant of how much the programme can actually accomplish.

Three questions for reflection

1. Think of your most important cross-boundary relationship right now. Which type of trust, competence, communication, or contractual, do you believe is most uncertain for the other person, and what have you actually done to address that specific type?
2. Is there a breach in any of your key cross-boundary relationships, inherited or recent, that has not been directly acknowledged? What is it costing the relationship every month that it remains unaddressed?
3. In your most recent interactions with key cross-boundary stakeholders, what proportion of the contact was initiated by what you needed versus what they needed? What does that ratio tell you about how your orientation is likely to be perceived?

Inspiration

Covey, S.M.R. and Merrill, R.R. (2006) The speed of trust: The one thing that changes everything. New York: Free Press.

Edmondson, A.C. (1999) ‘Psychological safety and learning behaviour in work teams’, Administrative Science Quarterly, 44(2), pp. 350–383.

Maister, D.H., Green, C.H. and Galford, R.M. (2000) The trusted advisor. New York: Free Press.

Mayer, R.C., Davis, J.H. and Schoorman, F.D. (1995) ‘An integrative model of organizational trust’, Academy of Management Review, 20(3), pp. 709–734.

Meyerson, D., Weick, K.E. and Kramer, R.M. (1996) ‘Swift trust and temporary groups’, in Kramer, R.M. and Tyler, T.R. (eds.) Trust in organizations: Frontiers of theory and research. Thousand Oaks, CA: Sage, pp. 166–195.

Reina, D.S. and Reina, M.L. (2006) Trust and betrayal in the workplace: Building effective relationships in your organisation. 2nd edn. San Francisco: Berrett-Koehler Publishers.