We were going to launch the feature on Tuesday.
Everyone on the team knew it wasn’t ready. The code was buggy. The UX was confusing. We hadn’t tested the edge cases. One of our engineers literally said in standup, “I’m not sure this is going to work well,” but immediately followed it with, “but I guess everyone else thinks it’s fine.”
The PM wanted to hit the deadline. The CEO was excited about the demo. The team had momentum. So we all nodded along.
We launched on Tuesday.
It crashed spectacularly. Customer support got flooded. We had to roll back in three hours. The CEO was furious. And in the post-mortem, five different people said, “I knew it wasn’t ready, but I didn’t want to be the one to say something.”
That’s groupthink. And it’s the invisible force that turns smart, experienced teams into disaster factories.
What is Groupthink?
Groupthink is a psychological phenomenon where the desire for group consensus overrides people’s common sense and critical thinking.
The term was coined by psychologist Irving Janis in 1972, after studying historical fiascoes like the Bay of Pigs invasion, the Challenger disaster, and Pearl Harbor.
His finding: Highly cohesive groups, under pressure to make decisions, often suppress dissent and make objectively terrible choices that individual members would never make alone.
The pattern:
- A group faces a decision
- A consensus emerges (often around the leader’s preference)
- Dissenting voices are subtly (or overtly) discouraged
- The group reinforces itself: “We’re all smart people. If we all agree, it must be right.”
- Critical thinking shuts down
- The group makes a terrible decision
- Everyone is shocked when it fails
The cruel irony? The more cohesive and intelligent the team, the more susceptible they are to groupthink.
You’d think smart people would be immune. Wrong. Smart people are excellent at rationalizing bad decisions that the group has already committed to.
The Symptoms of Groupthink
Janis identified eight classic symptoms. I’ve seen every single one in tech teams:
1. Illusion of Invulnerability
What it is: “We’re too smart/experienced/talented to fail.”
How it sounds:
- “Our team has never shipped a bad feature.”
- “We’re engineers from Google/Facebook/Netflix. We know what we’re doing.”
- “This will obviously work. Look at our track record.”
This excessive optimism leads to taking ridiculous risks.
Example: I worked with a startup that decided to rewrite their entire platform in a new language. Why? “We’re talented engineers. We can do it in three months.”
They were talented. They also underestimated by 9 months and nearly ran out of money.
Their talent made them overconfident. Classic groupthink.
2. Collective Rationalization
What it is: The group collectively dismisses warnings and negative feedback.
How it sounds:
- “Those users just don’t understand our vision.”
- “That’s an edge case. We can ignore it.”
- “Analytics are misleading. Our intuition is right.”
Example: A product team I consulted for was building a feature. User testing showed people were confused. Five different test participants couldn’t figure it out.
The team’s response? “They just didn’t give it enough time. Once people learn it, they’ll love it.”
They shipped it. Adoption was 3%. They were shocked.
They’d collectively rationalized away clear evidence that the feature was bad.
3. Belief in Inherent Morality
What it is: “We’re the good guys. Therefore, our decisions are good.”
How it sounds:
- “We’re doing this for the users.”
- “Our intentions are pure, so the outcome will be positive.”
- “We’re not like those other companies.”
This lets teams ignore ethical concerns or negative consequences.
Example: A company I knew decided to implement aggressive growth tactics—dark patterns, manipulative notifications, questionable data collection.
When engineers raised concerns, the response was: “But we’re trying to help people learn languages / stay healthy / connect with others. We’re the good guys.”
Their belief in their moral mission blinded them to the unethical means.
4. Stereotyped Views of Out-groups
What it is: Dismissing critics or outsiders as stupid, biased, or evil.
How it sounds:
- “Our competitors just don’t get it.”
- “Users who don’t like this feature are just resistant to change.”
- “That critic has an agenda.”
Example: A startup ignored critical feedback from industry veterans because “they’re old school and don’t understand modern SaaS.”
Turned out, the veterans were right. The startup’s pricing model was broken. But they’d stereotyped critics as “out of touch,” so they didn’t listen.
5. Direct Pressure on Dissenters
What it is: The group actively discourages dissent.
How it sounds:
- “Come on, be a team player.”
- “We’ve already decided this. Let’s move forward.”
- “Are you with us or not?”
This doesn’t have to be aggressive. Often it’s subtle social pressure.
Example: In a product meeting, someone says, “I have concerns about this approach.”
Someone else responds, “We’ve been over this already. Let’s not rehash.”
The dissenter backs down. The group moves on. The concern gets ignored.
6. Self-Censorship
What it is: People suppress their own doubts to avoid disrupting group harmony.
How it sounds (in your head):
- “Everyone else seems confident. Maybe I’m wrong.”
- “I don’t want to slow down the team.”
- “I’ll sound stupid if I bring this up.”
This is the most insidious symptom because it’s invisible.
Example: In the story I opened with, five team members had doubts. None spoke up. Why? Each thought they were the only one with concerns.
Self-censorship meant we launched a broken feature.
7. Illusion of Unanimity
What it is: Silence is interpreted as agreement.
How it happens:
- Leader: “So we’re all on board with this approach?”
- [Silence]
- Leader: “Great, we have consensus.”
But the silence doesn’t mean agreement. It means:
- “I have doubts but I’ll keep them to myself” (self-censorship)
- “I disagree but I don’t want to be difficult” (social pressure)
- “Everyone else seems to agree, so I’ll go along” (conformity)
The result: A false consensus that leads to terrible decisions.
8. Self-Appointed Mindguards
What it is: Some team members take it upon themselves to “protect” the group from dissenting information.
How it happens:
- Filtering negative feedback before it reaches the team
- Dismissing critics without sharing their concerns
- “Handling” dissenters privately
Example: A team member tells the PM about serious user complaints. The PM doesn’t share it with the team because “they’re already stressed and this will just distract them.”
The PM thinks they’re helping. They’re actually preventing critical information from reaching decision-makers.
Real-World Groupthink Disasters in Tech
Let’s talk about some famous examples:
The Challenger Disaster (1986)
The night before the Challenger launch, engineers warned that O-rings could fail in cold weather. The temperature was forecast to be 26°F—far below safe limits.
NASA managers dismissed the concerns. Why?
- Pressure to launch on schedule
- Belief that NASA was too competent to fail
- Stereotyping of dissenting engineers as “overly cautious”
- Self-censorship from engineers who had doubts but didn’t speak up
Result: Seven astronauts died.
The Rogers Commission found that groupthink was a major factor. Smart people, under pressure, suppressed critical thinking and made a fatal decision.
Nokia and the Smartphone Revolution
Nokia dominated mobile phones in 2007. When the iPhone launched, internal teams warned that Nokia needed to change dramatically.
But Nokia’s leadership was trapped in groupthink:
- Illusion of invulnerability: “We’re the market leader. We know phones.”
- Collective rationalization: “The iPhone is expensive. It’s a niche product.”
- Stereotyped view: “Apple doesn’t understand the mobile market.”
They dismissed the threat. By 2013, Nokia’s phone business was sold to Microsoft for scraps.
Theranos
A board full of impressive people (George Shultz, Henry Kissinger, James Mattis) oversaw a company claiming to revolutionize blood testing.
Scientists raised concerns. Employees flagged impossibilities. Journalists investigated.
But the board had groupthink:
- Belief in inherent morality: “We’re saving lives.”
- Illusion of invulnerability: “Look at our prestigious board.”
- Pressure on dissenters: Employees who raised concerns were threatened legally.
Result: Total fraud. Billions lost. Criminal charges.
Knight Capital’s $440 Million Loss (2012)
A trading firm deployed new software without proper testing. Engineers had concerns but kept quiet.
The software had a critical bug. In 45 minutes, it generated $440 million in losses. The company nearly went bankrupt.
Post-mortem revealed: multiple engineers had doubts, but groupthink and pressure to ship led them to stay silent.
Groupthink in Everyday Tech Work
You don’t need disasters to see groupthink. It’s in everyday decisions:
The Architecture Decision Nobody Questioned
Your team is choosing between SQL and NoSQL for a new project.
The tech lead prefers NoSQL. They’re excited about it. They talk about scalability and flexibility.
You have concerns—your queries are relational, your team knows SQL better, and you don’t actually need massive scale.
But you don’t say anything because:
- The tech lead seems confident
- Everyone else is nodding along
- You don’t want to look like you “don’t get it”
Six months later, you’re fighting with eventual consistency issues and writing terrible workarounds. Everyone wishes you’d chosen SQL.
Why didn’t anyone speak up? Groupthink.
The Hiring Decision
You’re interviewing a candidate. They’re charismatic. The first interviewer loved them. The second interviewer says, “Great culture fit!”
You interview them. You have doubts. Their technical skills seem weak. Their answers are surface-level.
But in the debrief:
- “I thought they were great!” (Interviewer 1)
- “Strong hire!” (Interviewer 2)
- “They’ll fit in well.” (Interviewer 3)
You think, “Hmm, maybe I’m being too harsh. Everyone else liked them.”
You say, “Yeah, they seemed good.”
Hire made. Three months later, the person isn’t working out. In a private conversation, two other interviewers admit: “Yeah, I had doubts too, but everyone else seemed enthusiastic.”
The Scope Creep Nobody Stopped
Project kickoff. Estimated at 2 months.
Week 3: PM suggests adding a feature. “It’s small. Won’t affect timeline.”
You think, “This will definitely affect the timeline,” but everyone says “sure, no problem.”
Week 5: Designer wants a redesign. “Just a quick visual update.”
You think, “That’s not quick,” but the team says “yeah, let’s do it.”
Week 7: Stakeholder requests an integration. “Should be easy.”
You think, “That’s a whole other system,” but the team is already planning it.
Month 4: Project is 2 months late. Everyone is surprised.
Nobody stopped the scope creep because nobody wanted to be the person saying “no” when everyone else was saying “yes.”
How to Recognize Groupthink in Real-Time
Here are the warning signs:
Warning Sign #1: You Have Doubts But You’re Not Voicing Them
If you’re thinking, “I’m not sure about this, but everyone else seems confident,” you’re experiencing groupthink.
The question is: are you the only one with doubts? Or are five other people having the same thought and also staying silent?
Warning Sign #2: Decisions Happen Too Quickly and Too Easily
Real, complex decisions are hard. They involve tradeoffs, competing concerns, and difficult choices.
If your team reaches consensus on a complex decision in 10 minutes with zero debate, you’re probably in groupthink territory.
Healthy teams have healthy conflict. They debate. They surface concerns. They explore tradeoffs.
Easy consensus on hard problems = groupthink.
Warning Sign #3: Dissenting Voices Get Shut Down
Someone raises a concern: “What about X?”
Someone else responds: “We’ve already discussed that.” or “Let’s stay focused.” or “We don’t have time for this.”
When raising concerns is treated as being difficult or negative, you have a groupthink culture.
Warning Sign #4: The Leader’s Opinion Becomes the Team’s Opinion
The VP says, “I think we should use microservices.”
Suddenly everyone thinks microservices are the right choice.
If the team’s opinion shifts dramatically based on what leadership thinks, that’s groupthink.
Healthy teams have independent thinking that informs the leader’s decision, not rubber-stamps it.
Warning Sign #5: External Feedback Gets Dismissed
User testing shows confusion. Customers complain. Metrics are bad.
The team’s response: “They don’t understand our vision.” “Give it time.” “The data is misleading.”
When you’re collectively rationalizing away negative feedback, you’re in groupthink.
How to Combat Groupthink
You can build systems and practices that make groupthink less likely:
Strategy 1: Assign a Devil’s Advocate
The practice: In every major decision, assign someone to argue against the proposed plan.
Not just passively, but actively. Their job is to poke holes, surface risks, and argue the opposite position.
Important: Rotate this role. Don’t always make it the same person (they’ll get stereotyped as “the negative person”).
Example:
“We’re deciding between React and Vue. Alice, your job today is to argue against whatever the team is leaning toward.”
This forces critical thinking. It legitimizes dissent. It surfaces concerns that might otherwise stay hidden.
Strategy 2: Anonymous Feedback Before Discussion
The practice: Before discussing a decision, have everyone write down their opinion privately and anonymously.
Why it works:
- Prevents bandwagoning
- Surfaces true opinions
- Removes social pressure
- Shows whether there’s real consensus or just apparent consensus
Example:
Before a product decision, send a form:
- Do you think we should build Feature X? (Yes/No/Unsure)
- What are your concerns?
- What are you uncertain about?
Aggregate results. Share them. THEN discuss.
You’ll often find: what looked like consensus was actually 40% yes, 30% unsure, and 30% no-but-I-wasn’t-going-to-say.
Strategy 3: Pre-Mortem
The practice: Before starting a project, imagine it has failed spectacularly. Why did it fail?
Have everyone write down reasons. Share them.
Why it works:
- Legitimizes raising concerns
- Surfaces risks before they become disasters
- Forces critical thinking
- Gives permission to be pessimistic
Example:
“It’s six months from now. Our product launch was a disaster. What went wrong?”
Responses:
- “We didn’t test with real users.”
- “The architecture couldn’t handle the load.”
- “We shipped too late and competitors beat us.”
- “The UX was too complex.”
These are concerns people had but weren’t voicing. The pre-mortem surfaces them.
Strategy 4: Bring in Outside Perspectives
The practice: Include people from outside the team in important decisions.
Why it works:
- Outsiders aren’t invested in the group’s consensus
- They ask “dumb” questions that reveal hidden assumptions
- They’re not subject to the social pressure of the core team
Example:
Engineering team is making an architecture decision. Invite someone from product or customer success to the meeting.
They’ll ask questions like: “Why are we doing this? What problem does it solve?” that the engineering team stopped asking three meetings ago.
Strategy 5: Leader Speaks Last
The practice: In meetings about important decisions, the leader/manager doesn’t share their opinion until everyone else has spoken.
Why it works:
- Prevents anchoring (everyone aligning to the leader’s first opinion)
- Encourages independent thinking
- Reveals true team sentiment
Example:
Instead of: “I think we should go with Option A. What does everyone think?”
Do this: “What do you all think? I’ll share my thoughts after I’ve heard from everyone.”
If you speak first, you poison the well. People will align to your opinion (consciously or not).
Strategy 6: Require Written Dissents
The practice: Before major decisions, explicitly ask: “Does anyone disagree or have concerns? If so, please write them down.”
Make it safe and expected to dissent.
Why it works:
- Makes dissent normal, not deviant
- Creates a record of concerns
- Forces people to articulate vague doubts
Example:
Amazon’s “disagree and commit” culture. You’re encouraged to disagree, but once a decision is made, you commit fully.
This only works if dissent is truly safe and encouraged.
Strategy 7: Red Team the Decision
The practice: Have a separate team whose job is to attack the plan.
Why it works:
- Creates structural opposition
- Finds weaknesses before they become disasters
- Prevents overconfidence
Example:
Security teams do this naturally. A dev team builds a feature, security red-teams it to find vulnerabilities.
Apply this to product/business decisions too. Have a team try to find reasons the plan will fail.
Strategy 8: Track Decision Quality Over Time
The practice: Keep a decision journal. Write down: what you decided, why, and what you expected to happen.
Six months later, review. Were you right? Wrong? What did you miss?
Why it works:
- Provides objective feedback on decision quality
- Surfaces patterns in bad decisions
- Makes overconfidence visible
Example:
After a product launch:
- Expected: 1000 users in month one
- Actual: 100 users
- Why wrong: We assumed our existing users would adopt immediately. They didn’t. We should have done more user research.
This feedback loop makes groupthink visible and learnable.
The Leader’s Role in Preventing Groupthink
If you’re a leader, you have special responsibility. Your team is watching you.
Do: Reward Dissent
When someone raises a concern, thank them publicly.
“I really appreciate Alice pushing back on this. This is exactly the kind of critical thinking we need.”
Even if you disagree with the concern, reward the behavior of voicing it.
Do: Admit Uncertainty
“I’m not sure this is the right call, but here’s my current thinking.”
This gives permission for others to be uncertain too.
Do: Change Your Mind Publicly
“Last week I thought X. After hearing your feedback, I’ve changed my mind. We’re doing Y.”
This models that changing your mind based on evidence is good, not weak.
Do: Create Psychological Safety
People won’t speak up if they fear negative consequences.
Make it safe to:
- Disagree
- Admit mistakes
- Ask “dumb” questions
- Say “I don’t know”
If your team is silent in meetings, it’s not because they have nothing to say. It’s because they don’t feel safe saying it.
Don’t: Punish Dissent
If someone raises a concern and you respond with “Let’s take this offline” or “We’ve already decided this,” you’re teaching the team that dissent is unwelcome.
They’ll stop raising concerns. And you’ll have created a groupthink machine.
Don’t: Have All the Answers
If you, as the leader, always have a strong opinion, your team will defer to you.
Sometimes, say: “I genuinely don’t know. What do you all think?”
Let them form opinions without your anchor.
When Consensus is Actually Good
Important caveat: Not all consensus is groupthink.
Healthy consensus:
- Comes after robust debate
- Includes diverse viewpoints
- Is based on evidence and reasoning
- Allows for dissent along the way
- Acknowledges tradeoffs and risks
Groupthink consensus:
- Comes quickly and easily
- Suppresses dissent
- Is based on social pressure and conformity
- Dismisses concerns
- Assumes there are no downsides
The difference: Did you have a real debate, or did everyone just nod along?
Advanced Technique: The “Dissent Percentage”
Here’s a framework I use:
For any major decision, I want at least 20-30% dissent or uncertainty.
If 100% of the team immediately agrees, I’m suspicious. Either:
- The decision is truly obvious (rare)
- People aren’t thinking critically
- People aren’t voicing concerns
- We’re in groupthink
So I’ll explicitly ask: “This seems too easy. What are we missing? What concerns do we have?”
I keep asking until I get meaningful pushback. If I can’t get any, I’ll delay the decision and assign someone to be a devil’s advocate.
100% consensus = red flag, not green light.
Personal Stories: My Groupthink Failures
Let me share times I’ve failed:
The Redesign Nobody Wanted
I was leading a product team. We decided to redesign our app. The design team was excited. The engineers were building it. Everyone was on board.
But we never actually asked users if they wanted it.
When we launched, users hated it. Complained loudly. Adoption dropped.
In retrospect, there were warning signs:
- User testing was lukewarm
- Support team had concerns
- Analytics showed people weren’t struggling with the current design
But we were so bought into the vision that we rationalized away the signals. Classic groupthink.
We eventually rolled back the redesign and learned a painful lesson about validating assumptions.
The Hire I Should Have Vetoed
I was interviewing a VP candidate. Charismatic. Great resume. The CEO loved them.
I had doubts. Their answers were rehearsed. They hadn’t done the specific role we needed. Something felt off.
But I thought, “The CEO has way more experience than me. They must see something I don’t.”
I said yes. The person was hired. Three months later, it was clear they weren’t working out. They left after six months.
Later, two other interviewers admitted: “Yeah, I had doubts too, but I didn’t want to be the one to say no when everyone else seemed enthusiastic.”
We all had doubts. None of us spoke up. Groupthink.
The Anti-Groupthink Mindset
Here’s what I’ve learned:
1. Dissent is a service to the team
You’re not being difficult. You’re not being negative. You’re helping the team make better decisions.
Voicing concerns is how you add value.
2. Easy consensus is suspicious
Hard problems don’t have easy answers. If a decision feels too smooth, dig deeper.
3. Your gut is data
If something feels off, that’s information. Don’t dismiss it because you can’t articulate it yet.
Say: “I have a concern I can’t quite articulate yet. Give me time to think it through.”
4. The team isn’t always right
Smart people make dumb decisions in groups. Trust your own judgment even when it contradicts the team.
5. Be the voice of dissent
Even when it’s uncomfortable. Even when everyone else agrees. Even when you might be wrong.
The cost of being wrong and speaking up is low. The cost of being right and staying silent can be catastrophic.
Final Thoughts: The Power of Productive Conflict
Groupthink happens when teams prioritize harmony over truth.
But the best teams don’t avoid conflict. They embrace productive conflict.
They debate. They disagree. They challenge each other. And through that friction, they make better decisions.
Harmony is comfortable. Truth is valuable.
I’d rather work on a team where we argue and make good decisions than a team where we all get along and make terrible ones.
So the next time you’re in a meeting and everyone is nodding along, ask yourself:
“Are we thinking critically, or are we just agreeing?”
And if it’s the latter, be the person who breaks the silence.
Your team will thank you. Eventually.
Have you experienced groupthink on your team? How did you handle it? What strategies have worked for you? I’d love to hear your experiences.