In 1906, Francis Galton attended a livestock fair in Plymouth where 787 people paid sixpence each to guess the weight of an ox. Galton — an elitist who believed expertise should determine policy — expected the crowd's guesses to be wildly inaccurate. He was wrong. The median guess was 1,207 pounds. The actual weight was 1,198 pounds. The crowd was off by less than 1%. No individual expert came closer. The statistician who trusted expertise over democracy had accidentally discovered the most powerful argument against expertise as the sole basis for judgement.
James Surowiecki formalised the insight in The Wisdom of Crowds (2004), identifying four conditions under which collective judgement outperforms individual experts: diversity of opinion (each person holds private information, even if it is only an eccentric interpretation of known facts), independence (each person's opinion is not determined by the opinions of others), decentralization (people draw on local knowledge and specialised expertise), and aggregation (a mechanism exists for turning individual judgements into a collective answer). When all four conditions hold, the errors in individual judgements cancel out and the remaining signal converges on truth with startling precision. When any condition fails, the crowd becomes a mob — and mobs are reliably worse than individuals.
Prediction markets are the cleanest modern demonstration. The Iowa Electronic Markets have outperformed major polls in predicting US presidential elections in 15 of 17 cycles since 1988. Prediction markets work because they satisfy all four conditions: diverse participants (traders with different information, different analytical frameworks, different biases), independence (each trader places their own bet without coordinating), decentralization (traders operate from different locations with different information sources), and aggregation (the market price is the mechanism that synthesises individual judgements into a single number). The price is not an opinion. It is the distillation of thousands of opinions, each weighted by the confidence the holder places in it — because in a prediction market, you back your belief with money.
Wikipedia demonstrates the same principle in knowledge production. The encyclopedia that anyone can edit has been shown in multiple studies to rival Encyclopaedia Britannica in accuracy across scientific topics. The mechanism is not that Wikipedia's individual editors are smarter than Britannica's experts. The mechanism is that the sheer diversity and volume of editors — each catching different errors, adding different knowledge, correcting different biases — produces a continuously self-correcting document that converges on accuracy through aggregation. The process looks chaotic from the outside. The output is reliable because the conditions are met: diverse editors, largely independent judgement, decentralised expertise, and a platform that aggregates contributions into a coherent article.
The critical insight — and the one most people miss — is that the conditions are demanding. Most "crowds" are not wise. They are herds. When diversity breaks down (everyone reads the same sources, follows the same influencers, shares the same priors), the crowd loses the error-cancelling diversity that makes it intelligent. When independence breaks down (people observe and copy each other's judgements rather than forming their own), the crowd converges on the first prominent signal rather than on truth. When aggregation fails (there is no mechanism to synthesise individual opinions into a collective judgement), the diversity exists but cannot be harvested. Financial bubbles, social media pile-ons, and political polarisation are all examples of crowds that fail the conditions — and crowds that fail the conditions are not just unintelligent. They are systematically worse than the individuals who compose them, because the herd amplifies errors rather than cancelling them.
Section 2
How to See It
Wisdom of crowds reveals itself through accuracy that exceeds any individual contributor's knowledge — not occasionally, but systematically. The diagnostic: when a group's aggregate judgement consistently outperforms its best individual member, the crowd conditions are being met. When the group converges on a single voice's position, the conditions have failed.
Product Development
You're seeing Wisdom of Crowds when Amazon's customer reviews produce product quality signals that rival professional review organisations. No individual Amazon reviewer has the expertise of a dedicated product tester. But the aggregate of thousands of reviews — each reflecting different use cases, different expectations, different failure modes — produces a quality signal that captures dimensions no single reviewer would test. The star rating is a crude aggregation mechanism. The real wisdom lives in the distribution: a product with 4.2 stars and 10,000 reviews is a more reliable quality signal than a single expert review, because the crowd has tested the product across conditions that no expert protocol could simulate.
Forecasting
You're seeing Wisdom of Crowds when a company's internal prediction market outperforms its planning committee's forecasts. Google, HP, and Intel have all experimented with internal prediction markets where employees bet on product launch dates, sales targets, and competitive outcomes. The markets consistently outperform the official forecasts produced by senior leadership — because the market aggregates information from engineers who know the code isn't ready, salespeople who know the pipeline is soft, and frontline employees who see problems that executives do not. The market price reflects the organisation's actual knowledge. The planning committee's forecast reflects what the planning committee wants to be true.
Investing
You're seeing Wisdom of Crowds when the stock market's aggregate valuation of a company more accurately predicts long-term business performance than any individual analyst's price target. Individual analyst estimates are noisy and biased (conflicts of interest, anchoring to prior estimates, herding around consensus). The market price — the aggregation of millions of independent buy and sell decisions — cancels the individual biases and converges on a valuation that reflects the collective assessment of the company's future cash flows. The efficient market hypothesis is, at its core, a wisdom-of-crowds argument.
Organisational Decision-Making
You're seeing Wisdom of Crowds when Bridgewater Associates' "believability-weighted decision-making" produces investment decisions that outperform both individual portfolio managers and equally-weighted committee votes. Ray Dalio's system does not treat every opinion equally. It weights each person's input by their demonstrated track record in the relevant domain — a modified aggregation mechanism that preserves diversity and independence while adjusting for expertise. The result is a crowd that is smarter than a simple majority and smarter than any individual, because the aggregation mechanism captures both the error-cancelling benefits of diversity and the informational advantage of domain expertise.
Section 3
How to Use It
Using wisdom of crowds is not about polling groups and averaging answers. It is about designing systems that satisfy the four conditions — diversity, independence, decentralization, aggregation — so that collective intelligence emerges structurally rather than accidentally. The design challenge is preserving the conditions against the social forces that naturally destroy them: conformity pressure destroys independence, information cascades destroy diversity, centralisation destroys local knowledge, and poor aggregation mechanisms waste the intelligence the crowd generates.
Decision filter
"Before making a high-stakes decision based on my own judgement or a small team's consensus, I ask: could I aggregate a wider set of independent judgements on this question? If yes, I design a process that collects diverse, independent inputs and aggregates them — and I weight the aggregate more heavily than any individual opinion, including my own. The crowd's answer is not always right. But it is right more often than any single expert, and over a career of decisions, systematically outperforming the base rate is the only edge that compounds."
As a founder
Your customers are a crowd — and their aggregate behaviour is smarter than your product intuition. The founder who builds based on what they think customers want is making a single-expert bet. The founder who builds based on what customer behaviour data reveals is harvesting crowd wisdom. Usage data, A/B test results, support ticket patterns, and churn reasons are all aggregation mechanisms that synthesise thousands of individual customer judgements into actionable signals. The signal is not what customers say in interviews (interviews are contaminated by social desirability bias and poor self-awareness). The signal is what customers do when no one is watching — because behaviour is the most honest aggregation of preference.
The internal version: create mechanisms for independent input before making strategic decisions. Amazon's practice of requiring written memos before meetings, read in silence for the first fifteen minutes, is a crowd-wisdom design choice. Each person forms their independent reaction to the argument before hearing anyone else's opinion — which preserves the independence condition that group discussion destroys. Compare this to the standard meeting format where the most senior person speaks first and everyone else anchors to their position. The first format harvests crowd wisdom. The second format creates a herd.
As an investor
The market is a crowd, and fighting its aggregate assessment is the highest-risk strategy in investing. This does not mean the market is always right — it means that when you disagree with the market, the burden of proof is on you. The market's price reflects the aggregated judgement of millions of participants, each with different information, different analytical frameworks, and different biases. Your individual analysis, no matter how rigorous, is one signal. The market price is millions of signals aggregated. To justify a contrarian position, you need a specific, articulable reason why you have information or insight that the crowd's aggregation has failed to capture — not a general belief that you are smarter than the market.
The diagnostic for when crowd wisdom fails in markets: check the conditions. Are participants diverse (or has the market converged on a consensus narrative)? Are judgements independent (or are traders copying each other in a momentum cascade)? Is there a functioning aggregation mechanism (or has liquidity dried up, preventing price discovery)? When the conditions degrade — as they do in bubbles, panics, and illiquid markets — the crowd's price stops reflecting wisdom and starts reflecting mania or despair. The investor who can distinguish between a wise crowd and a herd has an edge that compounds across cycles.
As a decision-maker
Design your decision-making processes to harvest crowd wisdom from your organisation. The enemy is the HiPPO — the Highest Paid Person's Opinion — which destroys independence by anchoring the group to a single, high-status viewpoint. Counter it with structural safeguards: collect written inputs before meetings, use anonymous polling for contentious decisions, separate idea generation from idea evaluation, and weight inputs by domain expertise rather than organisational rank.
Ray Dalio's "idea meritocracy" at Bridgewater is the most systematic implementation: every person's input is weighted by their "believability" in the relevant domain, determined by their track record of being right on similar questions. The system preserves the crowd's diversity (everyone provides input) while improving the aggregation mechanism (inputs are weighted by demonstrated expertise rather than by seniority or confidence). The result is a decision-making process that outperforms both pure democracy (where uninformed opinions dilute expert knowledge) and pure hierarchy (where a single leader's biases go unchecked).
Common misapplication: Treating crowd wisdom as crowd consensus. Consensus — the process of discussing until everyone agrees — destroys the very conditions that make crowds wise. Discussion creates social pressure that erodes independence. Persuasive speakers dominate, reducing diversity. The group converges on a single position, eliminating the error-cancelling variation that makes aggregation powerful. The output of a consensus process is not the wisdom of the crowd. It is the opinion of the most persuasive person with the crowd's endorsement stamp. True crowd wisdom requires independent inputs aggregated mechanically — not discussed into agreement.
Second misapplication: Applying crowd wisdom to questions that require specialised expertise. The crowd is wise on questions where many people hold partial information that can be aggregated — market predictions, quality assessments, demand forecasting. The crowd is not wise on questions that require deep domain knowledge concentrated in a few specialists — surgical technique, quantum physics, constitutional law. Galton's crowd could estimate an ox's weight because everyone had experience with animals and weight. The same crowd would be useless for diagnosing the ox's disease — because that requires specialised knowledge that diversity of opinion cannot substitute for.
Section 4
The Mechanism
Section 5
Founders & Leaders in Action
The two leaders below built organisations that systematically harvest crowd wisdom through structural design rather than cultural aspiration. Both recognised that individual judgement — even expert judgement — is limited, biased, and overconfident, and both built mechanisms that aggregate many independent inputs into decisions that outperform any single mind. The critical distinction: they did not just consult groups. They designed systems where the conditions for collective intelligence are preserved by architecture rather than by good intentions.
Bezos built multiple crowd-wisdom mechanisms into Amazon's operating system. Customer reviews — launched in 1995 when the idea of letting customers publicly criticise products on your own retail site seemed insane — are a crowd-wisdom engine. No individual review is authoritative. But the aggregate of thousands of reviews produces a quality signal that outperforms any professional review service, because the crowd tests the product across use cases, durability conditions, and expectation levels that no controlled review environment can replicate. Bezos understood that the wisdom was in the aggregate, not in any individual voice — which is why Amazon invested in making the aggregation mechanism visible (star ratings, verified purchase badges, "most helpful" sorting) rather than curating which voices were heard.
Amazon's internal decision-making reflects the same logic. The six-page memo, read in silence at the start of meetings, is a crowd-wisdom design choice. Each attendee reads the same document independently, forms their own judgement, and then contributes to the discussion. The silence preserves independence — no one's reaction is influenced by the most senior person's visible response. The structured discussion that follows aggregates the independent assessments into a collective evaluation that is richer than any individual reading. Bezos designed the process to prevent the meeting from collapsing into a herd following the HiPPO — and the design works because it operates on the structural conditions rather than relying on participants to resist social pressure voluntarily.
The "working backwards" process extends crowd wisdom to product development. By requiring teams to write the press release and FAQ before building the product, Bezos forces independent assessment of the product's value proposition from the customer's perspective. When multiple teams evaluate the press release — each applying their own expertise, each identifying different gaps — the aggregate assessment is wiser than any single team's evaluation.
Ray DalioFounder, Bridgewater Associates, 1975–present
Dalio built the most systematic implementation of crowd wisdom in the history of corporate decision-making. Bridgewater's "idea meritocracy" is an institutional architecture designed to satisfy all four conditions simultaneously. Diversity: every person in the organisation provides input on decisions, regardless of rank or tenure. Independence: input is collected through structured tools — dot ratings, surveys, written assessments — before group discussion, preserving individual judgement from social contamination. Decentralization: people apply their own local knowledge and specialised expertise rather than deferring to a central authority. Aggregation: the "believability-weighted" system synthesises inputs by weighting each person's vote according to their demonstrated track record on the relevant type of question.
The believability-weighted system is the critical innovation. Standard crowd wisdom treats every input equally, which works when the question is simple estimation (ox weight) but fails when the question requires domain expertise. Dalio's system solves this by modifying the aggregation mechanism: a portfolio manager with a 20-year track record of outperformance on macro trades has their input weighted more heavily on macro questions than a junior analyst with two years of experience. The system preserves diversity (the junior analyst still provides input) while improving accuracy (the experienced manager's input has proportionally greater influence). The result: collective judgement that benefits from both the error-cancelling power of diversity and the informational advantage of expertise.
The system's most radical feature: Dalio's own opinions are subject to the same believability-weighting as everyone else's. In standard organisations, the founder's opinion is the decision. At Bridgewater, the founder's opinion is one input, weighted by his track record, aggregated with everyone else's inputs. The design prevents the most common failure mode of organisational decision-making — the HiPPO effect — by making it structurally impossible. No individual, regardless of seniority, can override the crowd's weighted assessment without explicitly documenting the override and accepting personal accountability for the deviation.
Section 6
Visual Explanation
Section 7
Connected Models
Wisdom of crowds sits at the intersection of statistics, social psychology, and organisational design. It depends on diversity for its error-cancelling power, fails through groupthink and herding when its conditions break down, and is operationalised through aggregation mechanisms that range from simple averaging to sophisticated prediction markets. The connections below map the models that make crowd wisdom work, the models that describe its failure, and the models that extend its principles to broader strategic contexts.
Tension
Groupthink
Groupthink is the antithesis of crowd wisdom — the process through which a group's desire for consensus destroys the diversity and independence that collective intelligence requires. Irving Janis documented how groups of individually intelligent people — Kennedy's Bay of Pigs advisors, NASA's Challenger decision-makers — made catastrophically bad decisions because the group's consensus-seeking dynamic suppressed the dissenting information that would have produced a wiser collective judgement. The tension is structural: the same groups that could be wise (if opinions were collected independently and aggregated mechanically) become systematically wrong when opinions are formed through discussion, persuasion, and conformity pressure. Groupthink is what happens when a crowd loses its conditions and becomes a herd.
Tension
Herding
Herding is crowd wisdom's evil twin — the process where individuals abandon their private information and follow the visible behaviour of others. In financial markets, herding produces bubbles (everyone buys because others are buying) and panics (everyone sells because others are selling). The mechanism is rational at the individual level: if you believe others have information you don't, copying their behaviour is a reasonable strategy. But at the collective level, herding destroys the diversity that makes crowds wise — because instead of thousands of independent judgements, the herd produces a single amplified signal that may be entirely wrong. The transition from wise crowd to herd is often invisible: the crowd looks the same from the outside, but the internal dynamics have shifted from independent judgement to imitative behaviour.
Reinforces
Diversity of Thought
Diversity of thought is the first and most critical condition for crowd wisdom. Scott Page's mathematical proof — that collective accuracy depends on diversity of approaches as much as on individual ability — provides the formal basis for why homogeneous expert panels underperform diverse groups of competent generalists. The reinforcement is bidirectional: organisations that value diversity of thought create the conditions for wise crowds, and the demonstrable superiority of diverse collective judgement provides the business case for investing in cognitive diversity. The most common threat: hiring for "cultural fit," which selects for people who think similarly and destroys the diversity that makes collective decision-making valuable.
Section 8
One Key Quote
"Under the right circumstances, groups are remarkably intelligent, and are often smarter than the smartest people in them."
— James Surowiecki, The Wisdom of Crowds (2004)
The qualifier — "under the right circumstances" — is the quote's most important phrase, and the one most people skip. Surowiecki does not argue that groups are always smarter than individuals. He argues that groups are smarter when four specific structural conditions are met — and that the conditions are fragile, easily destroyed by the social dynamics that human beings naturally produce in group settings. The insight is not "trust the crowd." The insight is "design the conditions, then trust the crowd."
The phrase "smarter than the smartest people in them" sounds paradoxical until you understand the mechanism. The crowd is not smarter because it contains smarter individuals. It is smarter because it aggregates information that no individual possesses. Each person holds a unique slice of reality — different information, different experiences, different biases. When those slices are aggregated independently, the aggregate captures more of reality than any single slice. The smartest person in the room knows more than anyone else — but still knows less than everyone combined. Crowd wisdom is not about the quality of the parts. It is about the architecture of the whole.
The operational implication for leaders: the question is not "who is the smartest person to make this decision?" but "how do I design a process that aggregates the organisation's distributed knowledge into a decision that outperforms any individual's judgement?" The former leads to hero-worship and single-point-of-failure decision-making. The latter leads to institutional intelligence that survives the departure of any individual — including the leader.
Section 9
Analyst's Take
Faster Than Normal — Editorial View
Wisdom of crowds is simultaneously the most validated and most misapplied concept in decision science. The evidence for crowd superiority over individual experts — in prediction markets, estimation tasks, and information aggregation — is overwhelming. The evidence for crowds being reliably catastrophic when the conditions fail — bubbles, panics, mob behaviour, groupthink — is equally overwhelming. The concept is not complicated. The conditions are not mysterious. Diversity, independence, decentralization, aggregation. Four requirements. The challenge is that every natural social dynamic — conformity pressure, status hierarchies, information cascades, desire for consensus — actively destroys the conditions. Preserving crowd wisdom requires fighting human nature. And most organisations are not willing to fight that hard.
The pattern I track most closely: the gap between how organisations claim to make decisions and how they actually make decisions. Every company says it values diverse perspectives. Every company says it welcomes dissent. Functionally, most companies make decisions the same way: the most senior person in the room states their opinion, and everyone else calibrates their input to be close enough to that opinion to appear supportive but far enough to appear independent. This is not crowd wisdom. This is a HiPPO with an audience. The diagnostic is simple: at what point in the meeting does the most senior person share their view? If it is at the beginning, the crowd has been destroyed. If it is at the end — after independent inputs have been collected — the crowd has a chance.
Amazon's silent memo reading is the most elegant crowd-wisdom mechanism I have encountered in corporate practice. Fifteen minutes of silent reading before discussion. No PowerPoint, no presenter framing the narrative, no opportunity for the most senior person to anchor the room before independent evaluation occurs. The design preserves independence with a structural constraint that does not depend on participants' willpower to resist conformity. Compare this to the standard board meeting, where the CEO presents a deck, controls the narrative, and solicits "feedback" that is actually ratification. The structural difference produces radically different information quality — and information quality is the input that determines decision quality.
The danger I warn against most frequently: confusing the wisdom of crowds with the consensus of crowds. These are opposites. Crowd wisdom emerges from aggregating independent, diverse, private judgements. Crowd consensus emerges from discussion, persuasion, and conformity pressure that eliminate independence and diversity. The output of wisdom is accurate. The output of consensus is socially comfortable. Most organisations pursue consensus and call it wisdom — which is why most organisations make mediocre decisions despite employing brilliant individuals. The brilliance is there. The aggregation mechanism destroys it.
Section 10
Test Yourself
The scenarios below test whether you can identify the structural conditions for crowd wisdom — and whether you can distinguish between a wise crowd (diverse, independent, aggregated) and a herd (homogeneous, interdependent, converging on a single signal). The diagnostic is not whether a group made a good or bad decision. The diagnostic is whether the process preserved or destroyed the conditions that make collective intelligence possible.
Is this crowd wisdom or crowd failure?
Scenario 1
A tech company asks all 200 engineers to independently estimate how long a major infrastructure migration will take, submitting their estimates through an anonymous form before any group discussion. The median estimate is 14 months. The project lead — a senior architect with twenty years of experience — estimated 8 months. The CEO asks: should we use the crowd's estimate or the expert's estimate?
Scenario 2
A venture capital firm holds a Monday meeting where partners discuss new investment opportunities. The most senior partner presents a deal they are 'excited about' and asks the table for feedback. Each partner offers supportive comments, raises minor concerns that are quickly addressed, and the group reaches unanimous agreement to invest. Six months later, the company fails. Post-mortem analysis reveals that two junior partners had significant concerns about the market size but did not raise them because the senior partner's enthusiasm set the tone.
Section 11
Top Resources
The crowd wisdom literature spans statistics, social psychology, economics, and organisational behaviour. Start with Surowiecki for the accessible synthesis, extend to Page for the mathematical foundation, deepen with Sunstein for the failure modes, and add Tetlock for the applied research on how individuals can approach crowd-level forecasting accuracy through specific cognitive practices. The field is unique in that it produces both the strongest argument for collective intelligence and the strongest argument against it — depending entirely on whether the conditions are met.
The book that synthesised a century of scattered research into a coherent framework. Surowiecki identifies the four conditions for collective intelligence, documents cases where crowds outperform experts, and catalogues the failure modes that transform wise crowds into foolish mobs. Accessible and engagingly written, the book provides both the theoretical framework and the practical examples needed to understand when and why collective judgement works. The essential starting point for the topic.
Page provides the mathematical proof that cognitive diversity — different mental models, different heuristics, different interpretive frameworks — improves collective problem-solving independently of individual ability. The diversity prediction theorem demonstrates formally that a group's collective error equals the average individual error minus the diversity of predictions, which means increasing diversity reduces collective error even when individual ability stays constant. Essential for anyone designing teams, committees, or decision-making bodies.
Sunstein examines both the successes and failures of collective knowledge production — from Wikipedia to prediction markets to deliberating groups. The book provides the most thorough analysis of when and why group deliberation fails, documenting how group polarisation, cascades, and conformity pressure systematically degrade collective judgement. Critical for understanding why simply putting smart people in a room and asking them to discuss does not produce crowd wisdom — and what structural interventions can prevent the degradation.
Tetlock's research on the Good Judgment Project demonstrates that certain individuals — "superforecasters" — can approach crowd-level accuracy through specific cognitive practices: breaking questions into components, updating beliefs incrementally, and maintaining calibrated confidence. The book bridges crowd wisdom and individual decision-making, showing how the principles that make crowds wise (diversity of perspective, updating on evidence, aggregation of signals) can be internalised by individuals through deliberate practice.
Dalio's operating manual for Bridgewater Associates provides the most detailed case study of crowd wisdom implemented as institutional architecture. The book documents the "idea meritocracy" — the system of believability-weighted decision-making, radical transparency, and structured dissent that makes Bridgewater's collective intelligence a structural property of the organisation rather than a product of any individual's genius. Essential for leaders who want to move beyond "we value diverse perspectives" and build systems that actually harvest collective intelligence.
Wisdom of Crowds — Collective intelligence requires four structural conditions. When all four hold, the crowd outperforms experts. When any fails, the crowd becomes a herd that amplifies error rather than cancelling it.
Reinforces
Information Cascades
Information cascades describe the mechanism through which crowd wisdom collapses. In a cascade, each person observes the previous person's action, infers information from it, and copies the action — even when their own private information contradicts it. The cascade produces rapid convergence on a single position that may be wrong, because each subsequent person's decision reflects the cascade's accumulated signal rather than their own information. Cascades explain why crowds go wrong suddenly rather than gradually: the first few decisions set a trajectory that subsequent participants follow, and the trajectory is self-reinforcing because each follower adds social proof that attracts the next follower. Understanding cascades is essential for designing aggregation mechanisms that prevent them — which is why prediction markets (where bets are private and simultaneous) outperform sequential polls (where each respondent can see previous responses).
Leads-to
Prediction Markets
Prediction markets are the most sophisticated operationalisation of crowd wisdom — mechanisms that aggregate diverse, independent judgements through a pricing system that weights each input by the confidence its holder places in it. The market price is the crowd's answer, and the market's track record — consistently outperforming polls, expert panels, and official forecasts — validates the theoretical prediction that well-designed aggregation mechanisms outperform individual experts. The evolution from Galton's simple averaging to modern prediction markets is an evolution in aggregation sophistication: both harvest crowd wisdom, but the market adds incentive alignment (bet your money), self-weighting (confident participants bet more), and continuous updating (prices adjust in real time as new information arrives).
Leads-to
Aggregation Theory
Aggregation Theory — Ben Thompson's framework for understanding how internet platforms create value by aggregating demand — extends crowd wisdom from decision-making to market structure. Platforms like Google, Amazon, and Netflix aggregate millions of individual preferences (search queries, purchase behaviour, viewing history) into recommendations and rankings that reflect the crowd's collective assessment of relevance, quality, and value. The aggregation is crowd wisdom applied at scale: each individual's behaviour contributes a signal, and the platform's algorithm aggregates millions of signals into an output (search rankings, product recommendations, content suggestions) that is wiser than any individual curator's judgement. The connection is direct: aggregation theory is wisdom of crowds applied to market intermediation.
Scenario 3
A consumer goods company uses an internal prediction market to forecast quarterly sales. Employees across functions — sales, marketing, manufacturing, finance — can buy and sell contracts on whether quarterly revenue will exceed the official forecast. Over eight quarters, the prediction market's price has been more accurate than the official sales forecast in seven of eight quarters. The VP of Sales argues that the market 'undermines the credibility of the sales team's forecasts' and requests it be discontinued.
