translate Automatic translation · View original
Psychological safety ≠ being soft: building demanding and healthy teams
Canal De Escalado Y Liderazgo
The confusion between psychological safety and comfort has a real cost. Amy Edmondson, professor at Harvard Business School and the leading researcher on this concept, warns about it: psychological safety is not about being nice or lowering standards. It is about giving people with different perspectives the freedom to speak candidly (Edmondson, 1999).
When Edmondson began her research in the 1990s by studying medical errors, she discovered something counterintuitive: the best teams reported more errors than low-performing teams. Not because they made more mistakes, but because they had created an environment where it was safe to acknowledge them and learn from them. The lowest-performing teams simply swept their mistakes under the rug.
This distinction is fundamental. Psychological safety does not eliminate demands; it eliminates the fear that prevents improvement. It means creating a climate where people can take interpersonal risks without fear of negative consequences for their image, status, or career.
What is psychological safety, really?
Edmondson (1999) defines psychological safety as "a shared belief held by members of a team that the team is safe for interpersonal risk-taking." These risks include:
- Admitting mistakes or gaps in knowledge.
- Asking questions that may seem "basic."
- Proposing unconventional ideas.
- Challenging the status quo or the decisions of superiors.
- Requesting feedback or help.
Project Aristotle at Google analyzed 180 teams over several years to identify what made some of them high-performing and others not. The main finding was clear: psychological safety was the most important factor, above team composition, individual skills, or organizational structure (Rozovsky, 2015).
But here is the nuance that many organizations overlook: the teams with high psychological safety in Google's study also had high standards. They were not complacent teams. They were teams that combined the ability to question with a demand for excellence.
The four-quadrant model
To understand this apparent paradox, Timothy Clark proposes in The 4 Stages of Psychological Safety a useful framework that distinguishes between psychological safety and performance standards. When we cross these two variables (high/low psychological safety and high/low standards), we get four types of cultures:
- Low psychological safety + low standards: culture of apathy. Nobody takes risks because it isn't worth it. The bar is so low that it doesn't matter much either.
- High psychological safety + low standards: comfort culture. This is the one that generates the initial objection. Pleasant teams where everyone gets along, but without real ambition or demands.
- Low psychological safety + high standards: culture of anxiety. The most toxic. A lot is demanded, but no one can admit they don't know something or need help. People hide problems until they explode.
- High psychological safety + high standards: learning culture. The quadrant we should aspire to. Excellence is expected, but the support and candor needed to achieve it are provided.
The key is to understand that psychological safety is not the vertical axis of this matrix. It is not a matter of choosing between safety or demands. It is about building both simultaneously.
How psychological safety strengthens demands
Far from weakening standards, psychological safety makes them achievable in a sustainable way. Let's see how:
- More honest and frequent feedback. In teams with low psychological safety, feedback is postponed until formal evaluations, when it is already too late to correct course. In psychologically safe teams, feedback is an ongoing conversation. Psychological safety allows direct challenge without it being perceived as a personal attack.
- Experiments and rapid learning. Organizations that encourage systematic experimentation consistently outperform their competitors. But experimenting requires accepting that many experiments will fail. Without psychological safety, teams stop proposing experiments or only propose those with guaranteed success, eliminating precisely the potential for learning.
- Transparency about real problems. High-performing teams have a culture where it is safe to admit problems early. This allows them to be resolved while they are still small, instead of being discovered in production or in external audits (Forsgren et al., 2019).
- Productive debate on important decisions. Teams that tolerate dissent make better decisions. But tolerating dissent requires psychological safety. Without it, people practice self-censorship and the team loses valuable perspectives.
Three antipatterns that destroy the equation
When organizations try to create psychological safety without understanding what it really is, they often fall into predictable traps:
- Antipattern 1: confusing safety with the absence of conflict. Some teams interpret psychological safety as "here we all get along and never argue." Lencioni, in The Five Dysfunctions of a Team, identifies precisely this as a dysfunction: the absence of productive conflict. Healthy teams have intense disagreements about ideas while maintaining respect for people.
- Antipattern 2: not giving direct feedback for fear of "damaging psychological safety." Psychological safety does not mean avoiding difficult conversations. It means having them in a way that the person knows the feedback comes from respect and a genuine desire to help them grow.
- Antipattern 3: tolerating systematic underperformance. If someone consistently fails to meet the team's standards and there are no consequences, the psychological safety of the rest of the team erodes. Why? Because other members see that their effort doesn't matter or that they will have to compensate indefinitely for someone else's low performance. Psychological safety requires fairness: clarity about what is expected and consistency in how gaps are managed.
- Antipattern 4: using "psychological safety" as an excuse to avoid unpopular decisions. In these contexts, any action that might cause emotional discomfort is automatically perceived as a threat to the team's safety. The result is a dangerous drift toward inaction. Roles that clearly aren't working are not changed. The scope of unviable initiatives is not reduced. Persistent underperformance is not addressed. People are not removed from the team when the situation requires it. Everything gets stuck under an apparently ethical argument: “we don't want to damage psychological safety.” The problem is that this is not psychological safety; it is conflict avoidance disguised as care.
Building the right quadrant: concrete strategies
So how do we create that space where demands and safety are possible at the same time?
- Define clear, co-created standards. Ambiguity about expectations is the enemy of both safety and demands. The best teams take time to explicitly define what "excellence" means in their context. Not as an imposition from above, but as a conversation in which those who will do the work participate. This includes definitions of "done," quality criteria, working agreements, and collaboration expectations.
- Model vulnerability from leadership. Edmondson (2018) insists that psychological safety starts with those who hold the most formal power. When a leader admits they don't know something, that they made a mistake, or that they need the team's help, they establish that these behaviors are acceptable. It is not about pretending to be incompetent, but about authentic humility regarding the limits of individual knowledge.
- Separate the person from the behavior or result. There is a fundamental difference between "this design decision caused performance problems that we need to solve" and "you are a bad developer." The first is safe and demanding. The second is unsafe and destructive.
- Create structures for productive dissent. It is not enough to say "you can disagree here." Many people won't dare to without explicit structures that facilitate it. Some useful practices:
- Pre-mortems: before an important decision, the team imagines that it has failed and analyzes what could have gone wrong.
- Rotating devil's advocate: someone has the explicit responsibility of challenging the proposal.
- Temporal separation between divergence and convergence: first we generate options without judging, then we evaluate them.
- Celebrate learning as much as success. If we only celebrate when everything goes well, we are encouraging problems to be hidden. Psychologically safe teams have rituals for sharing what they learned from what didn't work.
- Manage underperformance directly and respectfully. When someone is not meeting the team's standards, the conversation should happen early and clearly. The model of difficult conversations by Stone, Patton, and Heen in Difficult Conversations proposes separating three aspects: what happened (observable facts), how we feel about it (emotions), and what it means about the relationship or identity (the most delicate conversation). Addressing all three with honesty is compatible with maintaining psychological safety.
The role of the Scrum Master and the leader
In Scrum contexts, the Scrum Master has a particular responsibility for cultivating psychological safety. Not as the "protector of the team against reality," but as a facilitator of difficult conversations that otherwise would not happen.
Some teams misinterpret the concept of a self-managing team, thinking it means "no one can question the team." The reality is the opposite: truly self-managing teams constantly question themselves. The Scrum Master helps those questions be productive, not destructive.
In retrospectives, for example, a common antipattern is that they become sessions where no one mentions real problems for fear of "pointing fingers." An effective Scrum Master creates structures that allow systemic problems to be discussed without making them personal. Techniques such as "Mad, Sad, Glad" or "Sailboat" help externalize problems, making them safer to address.
Systemic limitations
Not everything is within the reach of the team or the Scrum Master. If the organization has a punitive culture toward mistakes, it will be difficult to create psychological safety in an isolated team. If individual incentives are designed to encourage internal competition, real collaboration will be hard to achieve.
Ron Westrum, in his research on organizational cultures (2004), identifies three types: pathological (power-oriented, information as a weapon), bureaucratic (rule-oriented, information as a nuisance), and generative (mission-oriented, information as a vital resource). Psychological safety flourishes in generative cultures, but has to fight against the current in the other two.
Does this mean it isn't worth trying if we don't control the entire organization? No. It means we must be aware of our sphere of influence. As agile coaches, we can create pockets of psychological safety within teams, knowing that they will have limits in organizations with broader toxic cultures. And sometimes, the honest answer for a professional in that situation is to consider whether the organization deserves their energy in the long term.
The productive tension
Let's return to the initial question: can we have demanding and healthy teams? Not only can we: demands without psychological safety are unsustainable, and safety without demands is a waste of potential.
The tension between the two is not a problem to solve, but a paradox to manage continuously. The best teams move along that tightrope, constantly adjusting. There are moments when they need to tip the balance toward demands (a critical delivery, a commitment to a client) and others when they tip it toward care (after an especially hard sprint, when someone on the team is going through a difficult time).
The skill lies not in eliminating the tension, but in making it explicit and navigating it as a team. Are we demanding enough of ourselves? Are we caring enough about people? These questions never have permanent answers.
Psychological safety is not a state you reach and maintain effortlessly. It is a daily practice that requires intention, structures, and, above all, consistency between what we say and what we do as leaders and as teams.
What tensions do you observe in your team between safety and demands? Are there difficult conversations that are not happening? What small practice could you introduce tomorrow to make it a little safer to tell the truth?
Bibliography
- Brown, B. (2018). Dare to Lead: Brave Work. Tough Conversations. Whole Hearts. Random House.
- Clark, T. R. (2020). The 4 Stages of Psychological Safety: Defining the Path to Inclusion and Innovation. Berrett-Koehler Publishers.
- Dweck, C. S. (2006). Mindset: The New Psychology of Success. Random House.
- Edmondson, A. C.
- (1999). "Psychological Safety and Learning Behavior in Work Teams". Administrative Science Quarterly, 44(2), 350-383.
- (2018). The Fearless Organization: Creating Psychological Safety in the Workplace for Learning, Innovation, and Growth. Wiley.
- Klein, G. (2007). "Performing a Project Premortem". Harvard Business Review, 85(9), 18-19.
- Lencioni, P. (2002). The Five Dysfunctions of a Team: A Leadership Fable. Jossey-Bass.
- Scott, K. (2017). Radical Candor: Be a Kick-Ass Boss Without Losing Your Humanity. St. Martin's Press.
- Stone, D., Patton, B., & Heen, S. (2010). Difficult Conversations: How to Discuss What Matters Most (2nd ed.). Penguin Books.
- Thomke, S. H. (2020). Experimentation Works: The Surprising Power of Business Experiments. Harvard Business Review Press.
- Westrum, R. (2004). A typology of organisational cultures. BMJ Quality & Safety, 13(suppl 2), ii22-ii27.
Take part in the poll
This article includes an active poll in the Scrum Manager community.
poll View the poll in the community