A manager asks her team for candour and gets silence. She is competent, well-prepared, and right more often than not, so why won't anyone tell her the truth? The honest answer is rarely a character flaw. It is that trust is not one quality but several, and she is strong on some and quietly weak on one. The good news is that once you can name the parts, you can fix the part that's broken.

The quick version

  • Credibility is "can I believe your words?", your expertise and track record. It is the part most leaders over-invest in.
  • Reliability is "do you do what you said?", small kept promises, repeated. It is built slowly and broken in one missed commitment.
  • Rapport (the research calls it intimacy or benevolence) is "do you have my interests at heart, and is it safe to be open with you?" It is where most trust is actually won or lost.
  • The fastest way to destroy trust is to look self-interested. Be the most brilliant person in the room and it counts for little if people sense you're in it for yourself.

The idea in depth: trust has named parts

The most useful place to start is a deliberately simple formula from David Maister, Charles Green and Robert Galford's The Trusted Advisor (2000), the "trust equation." They express trust as (credibility + reliability + intimacy) ÷ self-orientation. Credibility is about your words, can you be believed? Reliability is about your actions, do you do what you say? Intimacy is the safety people feel being open with you. And the denominator, self-orientation, is how much you appear to be in it for yourself. It sits on the bottom of the fraction for a reason: it divides everything else. You can be credible, reliable and warm, and still not be trusted if people sense your real focus is you.

That reframes the work. Instead of "build trust" as one vague goal, you can audit yourself component by component. Ask, honestly: which of these four is my weak link? Most capable leaders are long on credibility and reliability and short on intimacy, and almost no one volunteers that their own self-orientation is the problem. If you rate yourself low on the bottom number, you have probably just found your blind spot.

The trust equation is a practitioner's model, not peer-reviewed science, so it earns its place by being memorable rather than measured. For the empirical backbone, leadership research points to the same idea from a different door. In their widely cited "An Integrative Model of Organizational Trust" (Academy of Management Review, 1995), Roger Mayer, James Davis and David Schoorman argue that we judge someone's trustworthiness on three factors: ability (competence at the task), benevolence (whether they have our interests at heart), and integrity (whether they hold to principles we find acceptable). Notice the overlap: ability maps onto credibility, integrity onto reliability, and benevolence onto rapport. Two independent frameworks, one practitioner and one academic, land in roughly the same place, which is the closest a soft topic gets to corroboration.

flowchart TD
  T(["Am I trusted?"]) --> C(["Credibility / ability
can your words be believed?"])
  T --> R(["Reliability / integrity
do you do what you said?"])
  T --> I(["Rapport / benevolence
is it safe to be open with you?"])
  C --> S{"Self-orientation:
am I in it for me
or for them?"}
  R --> S
  I --> S
  S -->|"low, for them"| Y(["Trust holds"])
  S -->|"high, for me"| N(["Trust collapses
whatever the other scores"])
The trust equation, redrawn: three things you build up, and one thing that divides them all. Leaders Loop

Why competence alone keeps failing you

The single most common mistake is assuming that being good at the job is enough. It rarely is. Edelman's annual Trust Barometer, a survey of tens of thousands of people across dozens of countries, has spent years separating two dimensions of trust: competence (do you deliver?) and ethics (are your intentions and conduct sound?). Its analysis found that across more than 600 companies, most score better on competence than on ethical dimensions such as integrity and purpose, and concluded that competence is not enough, that ethics-related factors weigh more heavily than ability in how much an organisation is trusted. The lesson scales down to a single leader: people will forgive you for not knowing something far sooner than they'll forgive feeling that you don't have their back.

The counter-intuitive part, for high performers especially: when trust is shaky, resist the urge to prove you're right. Adding more evidence of your competence pushes on the lever that's already strong. The slack lever is usually rapport, the felt sense that you understand the other person and are on their side. And it gets built in unglamorous ways. Asking a question and actually waiting for the answer. Remembering what someone told you last time. Putting the phone face-down so the person in front of you has your full attention.

Harvard's Frances Frei makes the same point with a slightly different anatomy. In her 2018 TED talk, drawn from a turnaround stint at Uber, she frames trust as resting on three legs, authenticity, logic and empathy, and argues that when trust wobbles, it's almost always because one specific leg is shaky. Her practical instruction is to find your own "trust wobble" rather than working on trust in general. Empathy wobbles, she notes, are often killed by distraction: a leader scanning a phone in a meeting is broadcasting that something else matters more than the people in the room.

People forgive not knowing something far sooner than they forgive feeling you don't have their back.

An honest limitation. These models are maps, not measurements. The trust equation has no validated scoring instrument, it's a thinking tool, and "rate your intimacy out of ten" is necessarily rough. The Mayer–Davis–Schoorman model is rigorous but describes trust between two specific parties in a specific context; it doesn't promise that being more benevolent in general earns trust from everyone, because trust also depends on the other person's disposition to extend it. And trust is culturally inflected: how much warmth versus competence a relationship needs before it deepens varies across teams and societies. Use these frameworks to diagnose, then test the diagnosis against what the specific person in front of you actually responds to.

A worked example

Return to the manager whose team won't speak up, call her Priya, leading an eight-person analytics team. (Illustrative scenario; not a real person.) On paper she should be the easiest person in the company to trust: deep expertise, hits her deadlines, never over-promises. Yet retrospectives are flat, and bad news reaches her late.

Run her through the components. Credibility: high, nobody doubts she knows the work. Reliability: high, she does what she says. Rapport/intimacy: low, she opens every one-to-one with the dashboard, not the person, and replies to problems by solving them before the other person has finished the sentence. Self-orientation: higher than she'd admit, under pressure she talks about how a slip makes her look to her boss. Her weak link isn't competence at all. It's the bottom of the fraction and the intimacy term, and piling on more competence has been making it worse.

flowchart LR
  A(["Symptom: team
won't speak up"]) --> B{"Which component
is the weak link?"}
  B -->|"credibility"| C(["High, she's expert.
Not the problem"])
  B -->|"reliability"| D(["High, she delivers.
Not the problem"])
  B -->|"rapport / self-orientation"| E(["Low / high,
this is the wobble"])
  E --> F(["Move: lead 1:1s with
the person, not the dashboard;
ask, then wait"])
  F --> G(["Bad news arrives
earlier, while it's
still cheap to fix"])
Diagnose before you treat, the obvious strengths weren't the problem. Leaders Loop

The fix is small and doable this week. Priya starts each one-to-one with a question about the person and holds the silence until they answer. When someone surfaces a problem, she asks "what do you need from me?" before offering a solution. She catches herself framing setbacks around her own standing and reframes them around the work. None of this is a personality transplant; it's a handful of deliberate behaviours aimed at the one component letting her down. Within a couple of cycles, the bad news starts arriving earlier, while it's still cheap to fix.

Frequently asked questions

What's the difference between trust, rapport and credibility?

They're nested. Credibility is one ingredient of trust, your believability, based on expertise and track record. Rapport is another, the felt connection and sense that you're on each other's side. Trust is the larger judgement that pulls them together with reliability (keeping your word) and low self-interest. You can have instant rapport with someone you don't yet trust, and deep credibility with someone you feel no rapport with. Lasting trust needs all the parts.

Can you build trust quickly, or does it only come with time?

Some of it is fast and some is slow. Rapport and credibility can form in a single good conversation. Reliability, by definition, can't be rushed, it's the record of promises kept, and one conversation can't prove it. That's why over-promising early is so costly: you spend credibility you haven't yet backed with reliability, and the gap is where trust breaks.

How do I rebuild trust after I've broken it?

Diagnose which component you damaged, because the repair differs. A reliability breach (you missed a commitment) is repaired by a visible run of small kept promises, not an apology speech. A rapport or self-orientation breach (people felt you put yourself first) is repaired by demonstrably acting in their interest, repeatedly, before you ask for anything back. Naming the specific wobble, as Frances Frei puts it, beats a generic charm offensive.

Isn't this just being likeable?

No, and conflating the two is a trap. Likeability is mostly rapport. Trust also demands reliability and credibility, which sometimes mean delivering news people don't want to hear. The most trusted leaders are often not the most universally liked; they're the ones who are warm and straight, who you'd want in the room precisely because they won't just tell you what's comfortable.

What's the single highest-leverage thing to work on?

For most capable professionals, it's lowering perceived self-orientation, the bottom of the trust equation. Because it divides everything else, a small reduction there lifts trust more than a large gain in any single numerator. In practice that means making conversations visibly about the other person's outcome before your own: ask first, solve second, and let people finish.

Related in the Toolkit

Trust is the visible output of work that starts inside you, managing your own reactions well (self-awareness & emotional self-regulation) is what keeps self-orientation low under pressure, and accurately reading what the other person needs (empathy & social awareness) is the engine behind rapport.

Where to go next