28 October 2026 · 14 min read

Lack of trust in the workplace: how it erodes, and what rebuilds it

Nobody has to betray anyone for trust to thin out. A date said aloud and not kept is enough for one person to start treating the next date as a maybe.

The notebook Orla keeps

Orla works in finance at a company of about eighty people. On Monday her manager, Declan, tells her he will have an answer on the contractor budget by Friday. He means it. On Friday he is in meetings from nine until six, and the answer does not come. On Tuesday, in the corridor, he says, "Still chasing it, sorry."

Nothing about this is a scandal. Declan has not lied, and the delay costs a week.

On Wednesday he tells the all-hands that the restructure will be settled "by the end of the month", and Orla hears a maybe. She writes the date in a notebook of her own, with a question mark. She starts a second plan in case it does not happen. At the next team meeting she keeps back the problem she was going to raise, because she has stopped assuming that what is said in this room turns into anything.

Declan, who has never broken a promise on purpose, does not know the notebook exists.

How much do employees trust their managers?

Declan is invented. The CIPD asked employees the question Orla is asking about him: 61% of employees with a line manager agreed that their manager can be relied upon to keep their promise, 19% neither agreed nor disagreed, and 16% disagreed (CIPD, Good Work Index 2026: Survey report, July 2026, Figure 37 on p. 35). It is one item about line managers, and the CIPD does not label it as trust.

On senior management, the same survey found that 85% of employees who trust their senior management to act with integrity were satisfied with their job, against 43% of those who do not; 78% would recommend their employer, against 23%; and 71% were unlikely to quit, against 48% (CIPD, same report, Figure 3 on p. 8). That is an association: people who are satisfied may be readier to trust.

In the US, just 20% of employees strongly agree they trust the leadership of their organization (Gallup, Global Indicator: Leadership and Management, "Trust in Leadership"). That is the strongest answer on the scale, not a count of everyone who distrusts, and the chart on the same page has read 19% or 20% at every reading from August 2024 to August 2026, so the line is flat.

None of these figures is a crisis, which is the uncomfortable part. A crisis gets a meeting. A middling score and a flat line get nothing, and the second plans go on being written.

Signs a team has stopped believing what it hears

Low trust rarely arrives as a complaint. It shows up as the extra work people do to protect themselves.

• A private tracker appears. Someone keeps their own version of the numbers or the dates, because the official one has stopped being enough.
• "Can you put that in an email?" A spoken yes used to do. Now people want it in writing.
• "Any questions?" gets silence. The real questions come in the corridor afterwards, where nobody who could answer them is standing.
• Updates read as all wins. The difficult part turns up at the review, and somebody in the room knew from the start.
• "They" means management. A theory about why something was decided goes round the lunch table until it sounds like a fact.

Any two of these together are worth taking seriously. Each one moves information away from where Declan could see it, and fewer questions and tidier updates can look to him like a team settling down. Distrust hides its own evidence, which is why the manager is so often the last to know.

What causes a lack of trust in the workplace?

Three judgements, made in small moments

Colquitt, Scott and LePine pooled 132 independent samples to ask what sits behind a feeling of trust. A trustee's trustworthiness, judged on ability, benevolence and integrity, and a person's general propensity to trust each explained variance in risk taking and job performance that trust itself did not, whether the trustee was a leader or a coworker (Journal of Applied Psychology, 2007). The studies are correlational. In plain words, and the gloss is mine: can they do it, do they wish me well, and will they do what they said.

None of this measures how fast trust goes. My reading is that the three judgements are made mostly in small moments, such as a date kept or missed, or a number given straight or rounded.

Notice which one Declan is losing. Orla does not doubt that he can do the job, or that he wishes her well. A capable, decent manager can still fail the third question, and capable, decent managers are the ones least likely to suspect it.

Why the nearest manager counts most

Dirks and Ferrin's meta-analysis of four decades of research on trust in leadership (k = 106) found that direct leaders, such as supervisors, appear to be a particularly important referent of trust (Journal of Applied Psychology, 2002). A direct leader's promises are also the ones a team gets to check inside a week.

Trust between colleagues is a separate account

De Jong, Dirks and Gillespie pooled 112 independent studies of 7,763 teams and found that trust among team members was positively related to team performance (ρ = .30), and that the relationship held after controlling for team trust in the leader and for past team performance (Journal of Applied Psychology, 2016). The ρ is a corrected correlation, how closely the two move together, not a share of performance explained. The link depended on how interdependent the work was, among other conditions, so the result is not that more trust always raises performance. Trust and performance can feed each other.

A silence does not stay empty

This part is my reading, not a finding. Orla has a date, no answer and no reason, so she supplies a reason of her own. The contractor budget is probably stuck in somebody's inbox. Left unexplained for long enough, it becomes a sign that the restructure has already been decided and nobody is saying.

What happens after trust breaks

What happens after trust breaks has its own evidence. In a laboratory exchange where players could pass money to a partner, Schweitzer, Hershey and Bradlow found that violated trust could be restored when people saw a consistent series of trustworthy actions, and that it recovered more slowly and less completely when the violation had involved deception (Organizational Behavior and Human Decision Processes, 2006, as reported by Knowledge at Wharton). A missed Friday is not deception.

The manager in the middle pays first

Follow Declan's Friday one floor up. The contractor budget is not his to sign. It sits with his director, Helen, who said "leave it with me" a fortnight ago and has been at a client all week. He gave Orla a date on somebody else's behalf, which is what most managers do most days, and he does not say so, because it feels like throwing Helen under a bus. So Orla gets "still chasing it, sorry", and the doubt lands on the person who said the date out loud.

That person is not in great shape either. Only 22% of managers worldwide were engaged in 2025, down from 27% a year earlier and nine points below 2022 (Gallup, State of the Global Workplace, 2026 edition, and our note on it). Engagement is not trust, and the figure says nothing about promises. It does describe the people who have been handed the job of keeping them.

What low trust costs a team

The work pays first, in hours nobody books. Orla's second plan is one job done twice. The problem she kept back at the team meeting is still a problem, nobody is working on it, and Declan is now managing from a picture with the bad news taken out.

The person pays too. In the CIPD's 2026 survey, 52% of employees who do not trust their leaders to act with integrity said their work has a negative impact on their mental health, against 15% of those who do (CIPD, same report, Figure 43 on p. 41). The CIPD reports that as a correlation and no more, so it does not show distrust doing the harm. It does show two groups of people having very different years at work.

Why trust falls and values statements do not rebuild trust

There are two easy wrong answers.

The first is the offsite with the trust fall. A trust fall asks whether a colleague will catch you, on a day when everyone is watching and nothing is at stake. The judgements above are formed from conduct over time, and a day away produces very little of it. A catch says nothing about whether Friday's date will hold. No study above tested an offsite, so this is my reading.

The second is the values statement. "We act with integrity" is a promise made for every manager at once, and it is checked the way any promise is checked, against Friday. It does not supply the track record the CIPD's question asks about. The first date that slips tests it.

How to build trust in a team, starting tomorrow

None of this needs a product. If you are asking how to build trust in a team, begin with the small things the judgements are made from.

List every date you have given in the last fortnight

Then keep each one or move it with notice. "I'll come back to you by Friday." For each, either do it, or go to the person before the date and give them a new one and the reason. A date that moves with notice is a different event from one that lapses and gets noticed.

Ask one person you manage what you said you would do and have not

Put it that way, privately: "Is there anything I said I'd do that I haven't?" Write the answer down without explaining. Then do one of the things on the list this week and tell them you did.

Say what is not decided, and when you will say more

The next time you pass a decision down, attach its one-line reason. Where something is not settled, or you cannot share it yet, say exactly that, with a date you can name. "By the end of the month" is a month wide.

The trouble with all three is the fortnight after, when the diary fills and an easy yes is quicker than a date you can keep. That is the usual death of a good intention, and the case for a small dose repeated over one big effort.

Trust at work: the questions managers ask

How do you rebuild trust at work, and how long does it take?

Through a consistent series of kept promises, and nobody has a reliable figure for how long. In the Schweitzer study trust could be restored that way, and it recovered more slowly and less completely after deception. So say what you promised, say what happened, give a new date, keep it, and keep the one after.

Should a manager apologise or defend themselves?

It depends on what the failure was about and on what the facts later show. In two studies of how trust is repaired after an alleged violation, trust recovered more when the mistrusted party apologised for a violation about competence but denied culpability for one about integrity, and when the response matched the evidence that came later: an apology if guilt was later shown, a denial if innocence was (Kim, Ferrin, Cooper and Dirks, Journal of Applied Psychology, 2004).

What can a team lead do if the distrust is of senior leaders?

Keep your own promises and pass on the reasons, because those are the parts your team can check. Dirks and Ferrin point to direct leaders as a particularly important referent of trust. Where a decision reaches you with no reason attached, ask for one before you pass it on, and where you cannot say, say that you cannot.

How MeaningFuel helps when trust is thinning

The daily insight: Trust, Integrity and Communication

MeaningFuel is the daily culture habit. A company chooses the values and skills it wants to work on, and each working day everyone gets one short insight on one of them, drawn from something a real person said. Eroding trust is a problem, and a problem cannot be chosen for the daily culture: that is what a module is for. What a company can choose is what the problem wears away: Trust and Integrity as values, and Communication as a skill.

In an ordinary week that might put a line on keeping your word in front of Orla on Tuesday, and one on saying what you do not know in front of Declan on Thursday. It will not get the contractor budget answered. It keeps those words in the working week, where the judgements are made, and off the all-hands slide.

The modules: Openness and Trust, and Trust in Leadership

Eroding trust has two modules because it runs in two directions: sideways, in how a team shares what it knows, and upward, between a team and its managers. A module is a short course of daily lessons: ten short lessons across two working weeks, one each working day. Whoever runs the account, or leads the team, books it for a team. It invites the team's named people, the lead among them, and those who join get one lesson each working day.

Openness and Trust is for a team whose updates arrive polished or late and whose real questions have moved into private messages. It draws on Ron Westrum's research on how organisations handle information, which the module reads this way: how freely information flows is a good sign of how much trust there is. In one lesson each person puts a number they would rather round up into their next update, said plain with no adjective in front of it.

Trust in Leadership is for the gap between the people who decide and the people who live with the decision, and both ends practise it. The team asks plainly in the room and brings a problem with two possible routes. The lead's half draws on Amy Edmondson's research on psychological safety, which gives the leader two jobs: inviting people to speak, and responding well when they do. One lesson gives the lead three honest answers: what I know, "I don't know yet, and here is when I will", and "I can't share that yet". "Still chasing it, sorry" is none of the three, because it carries no date. Another lesson sets up a visible list, kept by the team and not the lead: each question asked upward, who owns the answer, the date promised and the date answered. It is Orla's notebook, made public.

What leaders read, and what they cannot

Insights shows how much of a team takes part and how the insights land, pooled, with nobody named. For Declan's director that says whether the habit is alive in finance. It is not a reading of trust. It will not show the notebook, no screen shows how one person answered, and a team that goes quiet during a restructure is a team in weather, which is no verdict on its manager. During a module the lead sees whether the team is doing it, and never anyone's answers or what they write.

Both modules name their own limit. Where the problem is one particular broken promise, it needs a repair conversation, and the module does not replace it. What each does is turn the number said plain and the question answered by its date into habits a team keeps on a schedule, so they do not wait for somebody to remember. Neither promises a result, and neither will keep Declan's next date for him.

Back to Friday

Declan's missed Friday cost a week and one decision. Orla will put the notebook away when enough of his dates have held without her checking them, and the next one is the one to keep. That is what rebuilds it.

• Trust
• Lack of Trust
• Team Leadership
• Openness and Trust
• Trust in Leadership
• Workplace Culture