On August 14, 2019, the WeWork S-1 filing landed on the desks of analysts across Wall Street, and the reaction was unlike anything the tech industry had seen. Not because the numbers were bad. Because the numbers had been hiding in plain sight for years, and nobody had noticed.
Adam Neumann had raised $12.8 billion in private funding at a peak valuation of $47 billion. SoftBank's Masayoshi Son had committed $4.4 billion after a meeting that lasted twelve minutes. Twelve minutes. Jamie Dimon at JPMorgan had personally courted WeWork's IPO. Benchmark, T. Rowe Price, Fidelity, Goldman Sachs, Harvard's endowment. The most sophisticated investors on the planet had looked at a company losing $1.61 for every dollar of revenue and seen a technology company worth more than Ford, Delta, and Marriott combined.
Then the S-1 arrived. It opened with the phrase "We dedicate this to the energy of we," followed by a letter from Neumann that read more like a manifesto than a financial disclosure. The filing revealed that Neumann had personally borrowed $740 million against his stake, had trademarked the word "We" and charged the company $5.9 million to license it back, and had purchased buildings and leased them to WeWork at a profit. Revenue was $1.8 billion. Net losses were $1.9 billion. The company's own definition of "community adjusted EBITDA" excluded nearly every major expense, including the cost of building its communities. Within six weeks, the valuation dropped from $47 billion to approximately $8 billion. The IPO was pulled. Neumann was forced out.
The standard narrative frames this as fraud or delusion. But the more interesting question is the one nobody asked during the twelve-minute meeting: how did the most experienced stakeholders in global finance maintain alignment with a story that the numbers contradicted for years? The answer isn't stupidity or corruption. It's the neuroscience of how human brains process expectations, and why managing stakeholders has always been a problem that lives in the limbic system before it ever reaches the spreadsheet.
Stakeholder management is threat management. Every investor, board member, partner, and employee is running a brain that evolved to detect threats before opportunities, to anchor on first impressions, and to resolve conflicting information by protecting existing beliefs. The founders who manage stakeholders effectively aren't the ones with the best slide decks. They're the ones who understand that alignment is a neural state, and that maintaining it requires managing the specific brain circuits that produce trust, suspicion, and the uncomfortable middle ground between them.
The Threat-Detection Problem Nobody Talks About
In 1986, Joseph LeDoux at New York University published research that redefined how neuroscientists understood fear processing. LeDoux discovered that sensory information reaching the brain splits into two pathways. The "low road" routes directly from the thalamus to the amygdala, the brain's threat-detection center, triggering a response in roughly 12 milliseconds. The "high road" routes through the sensory cortex and prefrontal cortex for careful evaluation. The high road is more accurate. The low road is roughly ten times faster.
This architecture has a direct consequence for stakeholder management: your stakeholders form an emotional assessment of your credibility before their analytical brain has finished reading the first slide. The amygdala doesn't wait for the P&L statement. It scans your posture, your vocal tone, your eye contact, the pace of your speech, and the micro-expressions that cross your face when someone asks about runway. All of this happens below conscious awareness. By the time the analytical brain engages with the numbers, the emotional verdict is already in.
Neumann understood this intuitively. His physical presence was overwhelming. Six foot five, barefoot in meetings, radiating the kind of missionary certainty that the amygdala reads as "not a threat." His pitch wasn't about WeWork's unit economics. It was about elevating the world's consciousness, about a trillion-dollar addressable market, about being part of something unprecedented. The language was crafted for the amygdala's low road: big, certain, visionary, safe to follow. When Son committed $4.4 billion in twelve minutes, he wasn't evaluating a business model. His threat-detection system had cleared the signal, and the analytical pathway never fully engaged.
This is what happens in every stakeholder interaction, just usually at smaller scale. The investor who decides in the first ninety seconds whether to keep listening. The board member whose body language shifts before you've finished your sentence. The employee who reads your all-hands email and feels either reassured or alarmed before reaching the second paragraph. Every one of these reactions is the low road firing before the high road catches up.
A 2005 study by Alexander Todorov at Princeton demonstrated that people form judgments of competence and trustworthiness from faces in as little as 100 milliseconds. These snap judgments predicted the outcomes of U.S. Congressional elections with roughly 70 percent accuracy. The politicians didn't win because voters studied their platforms. They won because the amygdala cleared them as trustworthy before the prefrontal cortex opened the policy document. Stakeholder management operates on the same circuitry.
Why Do Stakeholders Resist Bad News (Even When They Asked For It)?
The S-1 revealed everything about WeWork that a rational analysis would have flagged years earlier. But the investors who had already committed billions didn't recalculate. Many doubled down. SoftBank tried to push the IPO forward even as the market recoiled. The psychology behind this isn't unique to WeWork. It's the same mechanism that kept Ron Johnson at J.C. Penney insisting customers needed to be "educated" while revenue fell by billions, and it has a name: cognitive dissonance.
Leon Festinger's 1957 theory describes the discomfort that arises when a person holds two contradictory beliefs. The brain doesn't tolerate contradiction. It resolves it, almost always by changing the belief that is least connected to identity. For an investor who has staked reputation and capital on a company, the belief "I made a brilliant investment" is identity-level. The belief "this company's numbers don't add up" is situational. When the two collide, the brain doesn't split the difference. It protects the identity.
Neuroimaging research by Keise Izuma and colleagues, published in PNAS, revealed the mechanism. When participants faced information that contradicted a prior decision, the dorsal anterior cingulate cortex (the brain's conflict-detection center) fired. The anterior insula generated aversive emotional arousal, the physical sensation of something being wrong. And the dorsolateral prefrontal cortex resolved the conflict not by rationally weighing the evidence, but by adjusting the belief that was easier to change. Participants inflated their evaluation of chosen options and deflated rejected ones, automatically, outside of conscious awareness.
For stakeholder management, the implication is uncomfortable: your stakeholders' brains are actively working to maintain their existing beliefs about you, your company, and the decisions they've already made. This is helpful when things are going well, because good news gets amplified by the same circuitry. But it becomes dangerous when problems emerge, because the brain's first response to bad news isn't evaluation. It's protection. The stakeholder doesn't hear "we missed our Q3 target by 15 percent." They hear a threat to the belief that they made a good decision investing in you, and the conflict-resolution machinery activates before the numbers finish loading.
This is why delivering bad news to stakeholders is so consistently difficult and so consistently botched. The founder who buries bad news in a dense update email, who sandwiches the revenue miss between two positive metrics, who waits until the quarterly board meeting to mention the problem that surfaced in week two, is not being deceptive. They are responding to a real signal: the stakeholder's brain will resist this information, and the messenger will absorb the emotional backlash of that resistance. The avoidance is rational. It's also corrosive, because every month of delay widens the gap between the stakeholder's mental model and reality, and the wider that gap grows, the more violent the correction when it finally arrives. WeWork's S-1 wasn't the problem. The problem was years of uncorrected drift between narrative and numbers.
The Anchoring Trap in Expectation Setting
Daniel Kahneman and Amos Tversky's anchoring research, first published in 1974, revealed that initial information disproportionately shapes all subsequent judgment, even when the initial information is arbitrary. In their famous experiment, participants spun a rigged wheel that landed on either 10 or 65, then estimated the percentage of African countries in the United Nations. Those who saw 65 guessed an average of 45 percent. Those who saw 10 guessed 25 percent. A random number had shifted their estimates by 20 percentage points.
Anchoring operates on every stakeholder interaction you will ever have. The first revenue projection you share becomes the anchor against which all future performance is measured. The first culture description you give a new hire becomes the standard against which every subsequent experience is judged. The first timeline you commit to a partner becomes the baseline that determines whether you are "on track" or "falling behind."
Neumann's anchor was a $47 billion valuation. Once that number existed in the minds of investors, employees, and the press, every subsequent piece of information was evaluated relative to it. An $8 billion valuation isn't objectively bad for a company with $1.8 billion in revenue. But measured against a $47 billion anchor, it felt like an 83 percent loss. The anchoring effect didn't just influence how people evaluated WeWork's worth. It determined whether the correction felt like a reasonable adjustment or a catastrophic collapse.
The practical consequence for founders is this: your initial anchor determines the emotional trajectory of every stakeholder relationship. Set it too high, and you create a gap that reality will eventually close with force. Set it too low, and you surrender the narrative energy that attracts capital and talent. The neuroscience suggests a specific approach. Because the amygdala processes threats more intensely than it processes rewards (the same loss aversion asymmetry that Kahneman documented at roughly 2:1), stakeholders feel a missed expectation roughly twice as painfully as they feel an exceeded one. A company that projects $8 million in revenue and delivers $10 million generates a moderate positive signal. A company that projects $12 million and delivers $10 million generates a threat response that is neurologically louder, even though both companies delivered the same result.
This is the math that most founders get backward. They anchor high to impress and spend every subsequent interaction managing the threat response created by the gap. The founders who maintain alignment over years do the opposite: they set anchors they can consistently exceed, training the stakeholder's brain to associate the relationship with reward rather than threat.
How Do You Deliver Bad News Without Triggering the Threat Response?
David Rock's SCARF model, published in 2008 in the NeuroLeadership Journal, identified five domains of social experience that the brain processes on the same neural circuitry as physical survival: Status, Certainty, Autonomy, Relatedness, and Fairness. A threat to any one of these domains activates the amygdala and prefrontal cortex in ways that mirror physical danger. A reward in any domain activates approach circuitry.
Bad news threatens at least three SCARF domains simultaneously. It threatens the stakeholder's certainty (their model of the future just changed). It threatens their status (if they championed the investment, their judgment is now in question). And it can threaten their autonomy (they may feel they've lost control of the situation). Three simultaneous threat signals is enough to push the amygdala past the point where the prefrontal cortex can maintain rational evaluation. This is why board meetings that start with bad news often devolve into emotional confrontation regardless of how well-prepared the presentation is.
The neuroscience suggests a specific sequencing. Leading with certainty, even partial certainty, dampens the threat response before the bad news arrives. "Here's what we know, here's what we've already done about it, and here's what we need to decide together" addresses certainty and autonomy before the loss is fully processed. Compare this to the more common approach: "We have some bad news. Revenue missed by 15 percent." The amygdala fires on "bad news." The loss is processed on "missed by 15 percent." By the time the founder gets to the plan, the stakeholder's brain is in threat mode, and threat mode doesn't evaluate plans. It evaluates threats.
Research on persuasion techniques confirms that the order of information presentation significantly affects how the information is processed. Presenting the response before the problem doesn't hide the problem. It gives the prefrontal cortex something to anchor on while the amygdala processes the loss. The stakeholder still hears the bad news. But they hear it in the context of a plan, which activates approach circuitry alongside the threat circuitry, rather than hearing it in a vacuum, which activates threat circuitry alone.
The best negotiation research supports the same principle. Chris Voss, the former FBI lead international hostage negotiator, has written extensively about the power of labeling the other person's emotions before presenting difficult information. "It sounds like this is going to be frustrating" activates the prefrontal cortex's emotion-labeling circuits, which research by Matthew Lieberman at UCLA has shown actually dampens amygdala activation. The label doesn't eliminate the threat. It gives the brain a cognitive frame to process it through, which is the difference between a stakeholder who listens to your plan and a stakeholder whose brain has already classified you as the threat.
Try This: The Stakeholder Alignment Audit
Most founders manage stakeholders reactively, waiting for tension to surface before addressing it. The neuroscience says this is exactly backward. By the time tension is visible, the amygdala has been firing for weeks and the cognitive dissonance machinery has been rewriting the stakeholder's model of your company. This protocol moves the intervention upstream.
Step one: list your five most important stakeholders. For each one, write down the specific mental model they hold about your company's trajectory. Not what you've told them. What they believe. If you're honest, you'll find gaps between the two, because the anchoring effect means their model is still partially anchored on whatever you first told them, regardless of what you've communicated since.
Step two: for each stakeholder, identify the single most important expectation that reality is likely to miss in the next ninety days. Revenue, timeline, hiring, product delivery, whatever it is. Calculate the gap between their current expectation and the most likely outcome. This gap is the cognitive dissonance your stakeholder will experience when reality arrives, and the width of the gap determines the intensity of the threat response.
Step three: close the gap before it closes itself. A 2015 study in the Journal of Experimental Psychology found that people who received incremental bad news showed significantly lower stress responses than those who received the same total bad news all at once. The brain can process small adjustments without triggering the full threat cascade. "We're tracking 5 percent below plan on Q3 revenue" in month one is a certainty adjustment. "We missed Q3 by 15 percent" in month four is a threat event. Same information. Different neural processing.
Step four: when delivering the update, lead with what you've already done. "We identified the gap three weeks ago, shifted spend from channel X to channel Y, and are now tracking to close 60 percent of the miss" addresses certainty and autonomy before the loss is fully processed. The stakeholder's brain receives the plan before it receives the problem, which keeps the prefrontal cortex engaged while the amygdala processes the adjustment.
Step five: after the conversation, send a written summary that re-anchors expectations to the new, more accurate number. The anchoring effect means the stakeholder's brain will drift back toward the original anchor unless you explicitly replace it. The written follow-up isn't administrative. It's neurological. It gives the brain a new reference point to calibrate against.
The WeWork collapse wasn't a failure of stakeholder management in the traditional sense. Neumann managed his stakeholders brilliantly for years. He set anchors that activated approach circuitry, crafted narratives that bypassed analytical processing, and created an identity-level commitment that cognitive dissonance then protected from contradictory evidence. The problem was that the management was pointed at perception rather than alignment. The gap between narrative and reality grew wider every quarter, and the correction, when it came, was proportional to the years of accumulated drift.
The founders who maintain stakeholder alignment over decades do something different. They manage the same neural systems, the amygdala, the anchoring circuits, the dissonance-resolution machinery, but they point them toward accuracy instead of inflation. They set anchors they can beat. They deliver bad news in increments that the brain can process without triggering threat cascades. They understand that every stakeholder is a brain running threat-detection software, and that the goal isn't to avoid triggering it. The goal is to give it accurate data, consistently, so that when the inevitable surprises arrive, the stakeholder's mental model requires adjustment, not demolition.
Chapter 7 of What Everyone Missed digs into the deeper framework on why the gap between perception and reality is the single most dangerous variable in any growing company, and why the founders who obsess over closing that gap outperform the ones who obsess over widening it.
FAQ
What is stakeholder management and why does it matter for startups? Stakeholder management is the process of identifying, communicating with, and maintaining productive relationships with everyone who has a vested interest in your company's outcome: investors, board members, partners, employees, and customers. For startups, it matters disproportionately because stakeholder misalignment can kill a company faster than a bad product. When a board member's mental model of your trajectory diverges from reality, the correction triggers threat-detection circuitry that can result in panic-driven decisions, withdrawn funding, or forced leadership changes. The neuroscience shows that stakeholders form emotional assessments of your credibility in as little as 100 milliseconds, and those assessments anchor every subsequent evaluation.
How do you deliver bad news to investors without losing their confidence? The neuroscience of threat processing suggests a specific approach: lead with what you know and what you've done before presenting the loss. David Rock's SCARF model shows that bad news simultaneously threatens a stakeholder's certainty, status, and autonomy. Presenting your response before the problem gives the prefrontal cortex an anchor while the amygdala processes the loss. Research also shows that incremental disclosure produces significantly lower stress responses than a single large revelation. Updating stakeholders early with "we're tracking 5 percent below plan" produces a manageable certainty adjustment; waiting until the quarter ends to reveal a 15 percent miss produces a threat event that activates fight-or-flight circuitry.
What is the connection between cognitive dissonance and stakeholder management? Cognitive dissonance is the discomfort that arises when a stakeholder holds two contradictory beliefs, such as "I made a smart investment" and "this company is underperforming." The brain resolves this discomfort by changing whichever belief is less connected to identity. For stakeholders who have publicly committed capital or reputation, the investment decision is identity-level, which means the brain will distort, dismiss, or reinterpret negative performance data to protect it. This is why investors sometimes double down on failing companies and why boards resist replacing founders they championed. Understanding this mechanism helps founders deliver bad news in ways that facilitate genuine recalibration rather than defensive denial.
How does anchoring bias affect stakeholder expectations? The first number, timeline, or projection you share with a stakeholder becomes the anchor against which all future performance is measured. Kahneman and Tversky's research showed that even arbitrary anchors shift subsequent estimates by 20 or more percentage points. Because loss aversion means stakeholders feel missed expectations roughly twice as intensely as exceeded ones, setting an initial anchor too high creates a neurological trap: every future update that falls short activates threat circuitry, regardless of whether the actual performance is objectively strong. The most effective founders set anchors they can consistently exceed, training the stakeholder's brain to associate the relationship with reward rather than threat.
Works Cited
- LeDoux, J. E. (1996). The Emotional Brain: The Mysterious Underpinnings of Emotional Life. Simon & Schuster.
- Todorov, A., Mandisodza, A. N., Goren, A., & Hall, C. C. (2005). "Inferences of Competence from Faces Predict Election Outcomes." Science, 308(5728), 1623-1625. https://doi.org/10.1126/science.1110589
- Festinger, L. (1957). A Theory of Cognitive Dissonance. Stanford University Press.
- Izuma, K., Matsumoto, M., Murayama, K., Samejima, K., Sadato, N., & Matsumoto, K. (2010). "Neural correlates of cognitive dissonance and choice-induced preference change." Proceedings of the National Academy of Sciences, 107(51), 22014-22019. https://doi.org/10.1073/pnas.1011879108
- Kahneman, D., & Tversky, A. (1974). "Judgment under Uncertainty: Heuristics and Biases." Science, 185(4157), 1124-1131. https://doi.org/10.1126/science.185.4157.1124
- Rock, D. (2008). "SCARF: A Brain-Based Model for Collaborating with and Influencing Others." NeuroLeadership Journal, 1, 1-9.
- Lieberman, M. D., Eisenberger, N. I., Crockett, M. J., Tom, S. M., Pfeifer, J. H., & Way, B. M. (2007). "Putting Feelings into Words: Affect Labeling Disrupts Amygdala Activity in Response to Affective Stimuli." Psychological Science, 18(5), 421-428. https://doi.org/10.1111/j.1467-9280.2007.01916.x
- "WeWork S-1 Filing." (2019). U.S. Securities and Exchange Commission. https://www.sec.gov/Archives/edgar/data/1533523/000119312519220499/d781982ds1.htm
- Neumann, A. (2019). WeWork S-1 Prospectus Letter. SEC Filing.