beyond the crunch: why high-performing teams need interdependence, not just pressure.

Randstad Advisory Signals from the Edge: this is a dispatch from your colleagues in 2030

High-pressure environments can create temporary bonds, but they often mask underlying systemic issues. For modern organizations, the goal shouldn't be to survive the next crisis, but to build teams that work together seamlessly by design. Learn about the science of team bonding, and why crunch culture isn't a substitute for real interdependence. Discover how to foster psychological safety, manage tacit knowledge and structure projects so that every team member can contribute. Find out what it takes to create a resilient culture where collaboration, not just individual effort, drives sustainable business results.

Think of the one person at work you would call first if something went badly wrong tomorrow — the colleague whose judgement you would trust at speed, without having to explain yourself.

Now ask when that trust was built. For most of us, the answer is some version of the same thing. The system that went down on a Friday afternoon. The audit nobody was ready for. The presentation with the impossible deadline. The shift where everything arrived at once. Whatever it was, it was hard, the outcome was in doubt and you came out of it knowing something about each other you had not known going in.

This is one of the most common experiences in working life and one of the least examined. Nearly everybody has a version of it. People often call this "trauma bonding." It is the wrong term, and the mistake is worth a moment.

Trauma bonding already means something much darker. Donald Dutton and Susan Painter used the term in 1981 to explain why people stay with partners who abuse them. When someone has power over you and switches between cruelty and affection, you become attached to them and the worse the treatment gets, the tighter the hold. It describes a trap. It has very little to do with four people getting a system back online at two in the morning.

But beyond misusing a clinical label, most organizations have very little to say about this shared-hardship phenomenon. That is a missed opportunity, because understanding it comes down to four critical questions:

1. 1. What actually produces strong employee bonds? 
   2. Where is it healthy and when does it become toxic? 
   3. Who is left standing outside the room while it forms? 
   4. And can any of this be built deliberately, even by teams who never share a physical space?

from ‘which kind of hard’ to ‘what the hard did to us’

There is a line between discomfort that builds a person up and stress that wears them down. The outcome of that stress depends far more on how well work aligns with someone’s strengths than on the sheer volume of the workload itself. While this is about a single individual and their role, we can also ask the same question of the space between people.

The intense period that erodes an individual can simultaneously produce the strongest working relationship of their career; same week, same person. 

Two things usually happen at the same time: the damage done to the individual, and the bond formed between them and whoever was in it with them. Most organizations only register one of the two. Those who see only the bond are romanticizing the crunch for what it built. Those who see only the damage done are stripping the pressure out, then wonder why nothing binds any more.

how are strong bonds formed?

This matters commercially. Gallup has spent three decades asking employees whether they have a best friend at work, and the item survives every attempt to cut it from the engagement index, because it keeps predicting performance, retention and safety. Close working relationships are an input to good work, rather than a byproduct of it.

Brock Bastian, Jolanda Jetten and Laura Ferris ran an experiment in 2014. A few groups did something painful together — they either put their hands in iced water, performed squats or ate hot chilli peppers — while other groups did the same tasks without the painful experiences. The groups that hurt together felt closer and cooperated more. Pain, apparently, was social glue. 

A follow-up study added a condition the original lacked: People experienced the same discomforts alone. The effects observed in shared experiences vanished. Experience sharing does the work, and it runs through empathy.

In an intimacy study, Arthur Aron showed that two strangers taking turns to answer increasingly personal questions could build, in thirty-six minutes, closeness that normally takes months. This was done with no adversity at all. Long nights bond people because somewhere around 11 p.m., still facing the uncertainty of not knowing the answer, people finally say something true about themselves. The crunch is often the only opportunity a workplace gives people to be this vulnerable and connect this authentically.

Two further conditions matter, and they turn up wherever strangers have to trust each other quickly. Debra Meyerson, Karl Weick and Roderick Kramer described “swift trust”: the trust that appears almost immediately in temporary teams working to a deadline on something genuinely uncertain — where people depend on each other and the consequences are real. They extend swift trust up front and check it as they go, before knowing each other at all. Film crews have it. Cockpit crews have it. 

Harvey Whitehouse's work on identity fusion explains why it lasts: When people look back on an intense experience as something that changed them, the line between who they are and who the group is blurs, and stays blurred.

So this bonding between colleagues is built in several ways: depending on each other, working without knowing the answer, facing real consequences and telling each other something true. 

High pressure earns its reputation honestly, because it is exceptionally good at forcing people into the trenches together to solve a problem. But it’s also not the only thing that can create that bond.

when pressure is healthy and when it is detrimental

There is a tipping point, and it is not marked by how hard the work feels. Three factors decide it:

1. 1. What kind of pressure it is: Michael LePine and Philip Podsakoff both found that demanding work raises motivation and performance, while obstructive work — bureaucracy, unclear roles and office politics, for example — only wears people down. 
   2. Agency and duration: Do people choose to take this on, and do they know when it will end? 
   3. Psychological safety: Is Amy Edmondson's concept of psychological safety present, allowing the team to say out loud what is going wrong while it is still going wrong? 

Psychological safety deserves more than third place on a list. The obvious assumption is that a hard week builds psychological safety, since people see each other at their worst and survive it. Edmondson's work points the other way: Safety is what allows a team to examine a failure rather than bury it. If so, psychological safety stops being just one factor out of three, and becomes the absolute precondition for all of it. The same difficult week bonds a team that already had psychological safety and damages one that did not.

The complication is that the bonding feels real on both sides of that line. Elliot Aronson and Judson Mills showed that the harder the initiation, the more people value what they were initiated into. A cohort surviving something brutal will sincerely believe the brutality made them, and hand it on to the next cohort as a gift.

This is how it becomes systemic. In February 2021, 13 first-year Goldman Sachs analysts ran their own survey of working conditions: 95-hour weeks, 5 hours of sleep each night. Self-rated mental health dropped down from 8.8 to 2.8, but employees also unmistakably described a bond with each other. 

Both things were true at once, but this doesn’t make the pain worth the benefit. In the same year, the WHO and ILO estimated that 55-hour weeks carried a 35% higher risk of stroke, and attributed 745,000 deaths.

The individual bond is defensible. The system that manufactures it as a rite of passage, and defends itself on the grounds of what it forges, is a different proposition entirely.

who gets left behind after high-pressure shared experiences?

Every high-pressure experience produces a second effect that rarely gets counted.

Three people come out of it knowing things they never wrote down because there was no time. Not the process, the other stuff: who to call, which risk was the real one, what the client actually meant. A fourth person joins afterward and walks into a room where a glance between two colleagues settles a question the new addition never even heard being asked.

As a knowledge problem, that is survivable. But this becomes a succession risk the moment those original three leave and take the undocumented knowledge with them. 

As an access problem it is more serious. The bond forms most reliably in conditions that are not available to a large portion of talent: caregivers, people with disabilities, those with visa constraints, for example. If an organization's deepest working relationships are made at 11 at night and miles away from home, it is giving in-office employees an unbalanced advantage. This happens with virtual teams too when the same select group is repeatedly invited to high-stakes collaborative meetings; they build the tacit knowledge and social capital that build critical bonds that can leave others out. 

This is where the design work starts, and three moves do most of the lifting.

1. 1. Run the review while the memory is warm.
      A blameless post-project review converts the tacit register into something a newcomer can read, and examining failure without blame is also how psychological safety gets built. Most organizations skip it exactly when it matters most, because the team is exhausted and the next project has already started.
      
   2. Rotate the high-stakes work.
      If the same people are always in the room when something matters, access to the bond is self-selecting, and networks compound accordingly. Varying who takes the difficult brief may cost a bit in delivery speed, but will build a far wider base of people who trust each other.
      
   3. Give newcomers a route to the tacit knowledge.
      Pair them with the people who hold the tacit knowledge, rather than the documentation that often does not. Tacit knowledge includes what was chosen, what was rejected and what caused concern. It carries more real knowledge than process documents ever do.

can strong bonds be built on purpose?

Partly, yes, but the evidence divides sharply on how.

Outdoor and adventure team-building, the version most often bought, is the least effective. A 2025 review by Tyne and colleagues found that the effects were small, measured on tests with little relationship to work and unlikely to survive the trip home.

Crisis simulation does it best: the medical and aviation tradition, crew resource management and TeamSTEPPS, where teams rehearse things going wrong while genuinely depending on each other. Steve Kozlowski draws the distinction that matters. It is not about how realistic the equipment looks; it is about whether the scenario makes people feel real uncertainty and real consequences. Though even then, staged emergencies produce weaker stress responses than real ones.

The most useful answer is less dramatic, and it is a redesign question rather than a training one. Real work already supplies uncertainty and consequences. What ordinary project structure strips out is interdependence. 

Companies typically design parallel workstreams that report upwards, resulting in people feeling like they don't need each other. They also strip out the unscripted time around the work, those unstructured moments where people decompress and say what they are actually feeling. Both can be designed back in during working hours. A shared deliverable two people cannot finish alone will do more than a day of team-building outings.

Remotely, it also works; it’s more challenging but it works. Swift trust was observed in distributed teams before anyone applied it to people in the same building, and its conditions need no room at all. What goes missing online is the unscheduled hour afterwards. Remote teams that bond schedule that time deliberately rather than waiting for chance.

questions worth asking to build strong employee bonds

The following questions are diagnostic rather than rhetorical. Each maps to a section above and has an answer you can find: in project histories, onboarding feedback, exit interviews or one honest conversation with a team that just went through something. When you cannot answer one, that gap needs exploring.

1. 1. On the stories people tell: Ask five colleagues to describe the work that gave them their closest working relationship. Count how many describe the pressure and how many describe the moment they told each other something true.
   2. On where it is healthy: Take your last intense work period ... Did people choose it? Did they know when it would end? Could they say out loud what was going wrong while it was still going wrong?
   3. On who is in the room: List the projects that produced your strongest teams over the past two years. Who could not have been on them, and what would have stopped them?
   4. On what you do afterwards: How many of your hard projects end with a structured debrief? How many simply end?
   5. On readiness: Would you describe your teams as safe enough to disagree openly? If the answer is no, you’ll need to fix that before anything else on this list.
   6. On what the group is holding: Does your last genuinely hard project exist as a written record or is it in three people's memories? What happens if those three leave within a year of each other?
   7. On design: When will people have to depend on each other most in the next six months, with real consequences and no obvious answers? Where have you built parallel workstreams that let people avoid each other entirely?

from hoping the bond happens to arranging the work so it does

The closeness people remember from the worst weeks of their careers was never produced by those worst weeks. It came from needing each other, not knowing the answers, facing something that mattered and saying something true. The crunch simply delivered all four at once, expensively, to whoever happened to be available. Those same conditions can be built into ordinary projects that finish at a reasonable hour.

Design the interdependence rather than hoping for it. Protect the debrief. Make the consequences of the work visible. Give new people a route to what the group already knows. Check that the team is safe enough for any of it to land.

This belongs to how work is organized, which is exactly why it tends to fall between the people who own culture and the people who own delivery.

If you want to work through where your organization's real bonds are being built, and who is not there when it happens, we are ready for that conversation. Following is a tool you can also use in the meantime.

high-pressure bonding: the cheat sheet

Most of us have a working relationship that came out of a bad week, and most of us credit the bad week for it. Yet what actually powers performance is something else. When people trust each other implicitly, they make high-stakes decisions at speed and solve problems without waiting for permission or shielding themselves from blame. This is what Randstad Advisory calls the “key 5” —  four conditions and one foundation that drive above-and-beyond performance.

the key 5: 4 conditions & 1 foundation

1. 1. Interdependence: People genuinely cannot finish without each other. 
   2. Uncertainty: The answer is not known at the start.
   3. Real consequences: Something is visibly at stake. 
   4. Reciprocal disclosure: People know something true about each other.
   5. The foundation: psychological safety. Where it is high, pressure builds something. Where it is low, the same pressure does damage.
      
      

4 things high-pressure bonding is not

• 1. Long hours: This predicts exhaustion well and closeness poorly. 
  2. Liking each other: Shared commitment to the task does more work than warmth.
  3. Relief at having pulled it off: The same week ending in failure gets remembered differently. 
  4. The off-site bonding day: This is the weakest evidence base of anything here.
     
     

5 questions to ask about your organization

1. 1. Where do two people genuinely need each other to finish, and where have you built parallel tracks that let them avoid each other?
   2. When people describe the work that made them close, do they describe the pressure or the conversation?
   3. How many hard projects end in a structured debrief, and how many simply end?
   4. Who missed out on the hard projects that built your team's strongest bonds? 
   5. Are your teams safe enough to disagree openly?
      
      

5 next steps 

1. 1. Design the interdependence: shared workflows, some goals and rewards at team level 
   2. Protect the debrief: eighteen minutes, structured and buys roughly a quarter improvement in performance 
   3. Make consequences and impact visible: people in contact with whoever their work affects 
   4. Build connection into arrival: new starters trained with the team and paired with the people holding the knowledge 
   5. Rotate the demanding work: equitable and intentional access to the best relationships