Enough with the Spaghetti Towers!
For years I worked inside organisations as an employee, and I lost count of the training days that started the same way. In would come a leadership facilitator, and the tables would already be set: bundles of uncooked spaghetti, a roll of masking tape, a length of string and a single marshmallow per table.
You know the drill. Eighteen minutes on the clock. Tallest freestanding tower wins. The marshmallow goes on top.
I understood what they were trying to do. I could see the intention, and I respected the effort someone had put into organising the day. I still walked out every time with the same feeling: it fell flat.
Eighteen minutes of laughter, then back to the silos
My table was never short on energy. Someone always appointed themselves chief engineer. Someone quietly held the base steady. Someone else cracked jokes when the whole thing buckled with two minutes to go. We laughed a lot. Once or twice we even won.
Then the facilitator ran the debrief, we wrote "communication" and "trust" on a flip chart, and everyone went back to work.
Back at our desks, nothing had moved. The team that held information close kept holding it close. The two managers who couldn't agree on anything still couldn't agree. The decision that had been stuck for months stayed stuck. People who had laughed together over a pile of spaghetti went straight back to working around each other.
We could cooperate. We had proven it before morning tea. What we couldn't do was cooperate when something we cared about was on the line, and no amount of spaghetti was going to change that.
Why the tower keeps getting booked
To be fair to the people who keep ordering it, the spaghetti tower has a lot going for it.
It's cheap. It needs almost no preparation and works for any group size. It generates energy quickly: competition, laughter, a visible winner. Failure is funny rather than costly, so nobody feels exposed. It photographs well, which lets an organisation show it has invested in its people.
It also comes with a respectable pedigree. The Marshmallow Challenge was popularised by Tom Wujec's 2010 TED talk, which carried a memorable finding: kindergarten children tend to build taller towers than business school graduates, because they start building straight away and learn from each collapse, while the graduates spend their time planning one perfect attempt.
The spaghetti tower also has plenty of relatives. LEGO® Serious Play®, blindfolded rope squares, escape rooms, raft building on the lake. Different materials, same promise: do something novel together and the bond will follow you back to the office. LEGO Serious Play has the most considered method behind it, and I understand the thinking. In my experience it lands in the same place, because the problem it's trying to solve sits outside the room.
The logic behind booking it goes something like this. Put people under time pressure with a shared goal, and they'll communicate, take on roles and solve problems together. Debrief it well, and they'll carry those lessons back to the office.
Twenty sticks of spaghetti and no history
The trouble is that these exercises remove every condition that makes teamwork hard in a real organisation.
There is no history. For eighteen minutes, the restructure never happened, the promotion went to the right person and nobody forwarded that email.
There are no competing priorities. Everyone wants the same tower. At work, the people at that table answer to different KPIs, different budgets and often different directors.
There is nothing scarce that matters. Nobody is asked to hand half their spaghetti to the next table so another team can hit its target.
There are no consequences. When the tower falls, people laugh. When a shared project falls over at work, someone wears it.
Power is flattened for eighteen minutes, then quietly restored the moment everyone sits back down.
So the activity tests the one thing most of your people already do well: cooperate when interests are aligned and the stakes are low. It avoids every situation where they struggle.
The research backs this up. A widely cited meta-analysis of team-building interventions (Klein and colleagues, 2009) found the strongest effects came from goal setting and role clarification, the unglamorous work of agreeing what a team is for and who is responsible for what. Activities aimed at interpersonal warmth did far less. Google's Project Aristotle reached a similar place from a different direction, finding that psychological safety, the belief that you can speak up without being punished, mattered more than who was on the team.
Even the kindergarten finding gets lost. Its lesson is about prototyping and learning from failure. In most debriefs I've sat through, it gets reduced to a line about communication and everyone heads to lunch.
Teams that don't work well together are telling you something
When teams in your organisation don't play well together, it's rarely because they haven't had enough fun with each other. Friction between teams is usually a symptom, and the cause sits somewhere less comfortable.
In my work with organisations, the same causes come up again and again:
Conflicting priorities set at the top. If the executive team hasn't genuinely agreed what matters most, each division protects its own, and the teams beneath them fight it out.
Unclear decision rights. When two teams both believe they own a decision, or both believe the other one does, you get either conflict or paralysis.
Incentives that reward the wrong thing. If people are measured on divisional results, they will optimise for divisional results, and collaboration becomes a favour rather than an expectation.
History that hasn't been dealt with. Broken promises, restructure fallout, a leader who once punished someone for speaking candidly. People remember.
Undiscussables. The underperformer protected by seniority. The director everyone works around. The project everyone knows is failing.
These things will never surface over a pile of spaghetti, because the activity is designed to feel safe. They are precisely what people won't say in a room with their manager present, unless the conditions make it safe and someone asks the right question.
There's one more pattern I see almost everywhere. Teams copy the team above them. If your executive group operates as a collection of portfolio holders defending their turf, your middle managers will do exactly the same, and they will do it with more conviction because they've watched you model it.
Measuring the conditions instead of the mood
Engagement surveys tell you how people feel. A pulse survey tells you whether that feeling has gone up or down since last quarter. Both have their place. Neither tells you why people feel that way, where the pressure is building, or what to do about it.
That gap is why I developed Culture Compass™. It looks across every major area of organisational life and maps an organisation across four dimensions: Vision, Aspiration, Capacity and Risk. Where those dimensions meet places the organisation in one of four quadrants: Competitive Advantage, Strategic Tension, Systemic Exposure or Credibility Tension.
The quadrant matters less than what sits behind it. Culture Compass™ is designed to show whether people believe the direction they've been given, whether they have the capacity to deliver it, and where risk is accumulating unnoticed. Most importantly, it shows the distance between what the executive believes is happening and what people across the organisation are actually experiencing.
The results aren't emailed out as a dashboard. They're worked through, discussed openly and presented back to the organisation as the foundation of an action plan. That's where the undiscussables start to get discussed, because the data has already put them on the table.
A morale score might tell you that 62 per cent of your staff are engaged. It won't tell you that two of your directorates have been quietly working against each other for eighteen months, or that your people leaders have stopped raising problems because nothing happens when they do. That is the information you need before you spend another dollar on team building.
Three things you can do today
1. Look back at your last executive meeting. Did anyone argue for something that would cost their own area? Did the real disagreements happen in the room, or in the corridor afterwards? Did decisions stick, or get relitigated a week later? If those answers make you uncomfortable, that is where your teamwork problem starts. Fix the team at the top before asking anyone else to work differently.
2. Find the handover that hurts the most. Ask three managers below executive level one question: where does work get stuck between your team and another team? Listen for the answer that comes up more than once. Then get the leaders involved in a room and agree who decides what. Role clarification is unglamorous, and it is where the research says the gains are.
3. Ask what the next team day is meant to fix. Before you sign off the next offsite, ask the person organising it what specific problem it will solve. If nobody can answer clearly, you don't yet understand the problem well enough to fix it. Spend that budget on finding out instead, through a proper diagnostic and a well-facilitated conversation about what people aren't saying.
The tower was never the problem
Your teams can build a tower out of spaghetti in eighteen minutes. If they can't build anything together for the rest of the year, the problem was never the teams.
