Data is Overrated
When Intuition Beats Data-Driven Decisions
Data is considered the new oil, the hard currency in an increasingly rationalized innovation system. Hardly a pitch that isn’t peppered with metrics. Hardly an accelerator that doesn’t demand data-driven validation. The motto: Only what can be measured actually exists. But what if this very logic blinds us to what really matters?
In startup reality — particularly in early phases — valid data is often an illusion. Markets don’t exist yet, user feedback is distorted, and seemingly hard KPIs lead to false conclusions. The pressure to substantiate everything with numbers tempts us to measure the wrong things, ask the wrong questions, or discard the right things too early.
Modern decision psychology shows: Intuition is not the opposite of rationality, but rather another form of intelligence. It’s based on unconscious pattern recognition, on condensed experiential knowledge, and is often superior in highly complex, uncertain situations. Psychologist Gerd Gigerenzer aptly describes this as “intelligent gut feeling” [1]: a concept that is underestimated, particularly in entrepreneurship. Gigerenzer’s research shows that people frequently make better decisions when they rely on so-called heuristics rather than complex models.
Daniel Kahneman [2] — often celebrated as the father of behavioral economics — also emphasizes that intuition in expert systems (such as with experienced founders) is often accurate, as long as they receive feedback. The blanket devaluation of intuitive decisions therefore falls short.
Data is important, but it’s not objective. It consists of constructs embedded in assumptions, models, and measurement errors. Intuition, on the other hand, recognizes patterns before they can be translated into KPIs. It’s particularly valuable where no historical data exists: in radical innovations, new markets, and profound societal transformations.
Moreover: Intuition is system-capable. While data thinks in linear causalities, intuition often recognizes non-linear relationships, interactions, and systemic tensions long before they appear in reporting. Anyone who wants to operate in complex systems needs the ability to perceive nuances, emergent dynamics, and implicit signals. That’s exactly what intuition delivers.
Those who rely exclusively on dashboards lose sight of the whole picture. Those who understand data as a tool and intuition as a strategic compass make more well-founded decisions. This article shows when which form of knowledge leads. And why founders should learn to balance both.
Data ≠ Reality: Why Numbers Don’t Always Count
Data is usually treated as hard facts, as reliable representations of reality. But this is a fallacy. Data is not reality, but always a selection, interpretation, and simplification. It only reflects what has been made measurable — not what is relevant. This discrepancy is particularly risky for founders: those who rely too heavily on numbers can overlook opportunities that cannot yet be captured in dashboards.
Numerous biases creep in already during data collection. Sampling bias occurs when the collected data is not representative of reality: for example, when only the loudest or most active users are surveyed. Confirmation bias leads teams to preferentially collect or interpret data that confirms their existing assumptions. And measurement errors slip in through imprecise question formulations, inconsistent metrics, or sample sizes that are too small.
A classic example is measuring product-market fit in early phases: When a startup asks users how disappointed they would be if the product no longer existed — as described in Sean Ellis’s popular PMF survey [3] — the answers are often only meaningful when a clear user group already exists. In new, emergent markets, however, there are often no stable expectations or comparison values. The numbers suggest certainty where uncertainty actually prevails.
Particularly in early phases, the most relevant information is often not quantifiable. Those who scale too early because an A/B test shows good numbers but the problem hasn’t been understood deeply enough are building on sand. Equally fatal: discarding radically new concepts with weak data backing, even though they address systemic pain points that are still slumbering beneath the surface.
Therefore: Not every measurable feedback is useful [4] and not every silent signal is meaningless. Intuition, conversations, pattern recognition, and systems thinking complement data where it doesn’t (yet) apply. And this happens more frequently in disruptive markets than many believe.
The Past as a Trap: When Data Blocks the Future
Data analyses are by definition based on the past. They show what has worked, not what will work. For startups that don’t just want to serve existing markets but create entirely new ones, this is a central problem. Those who rely too heavily on historical data during the innovation process optimize incrementally at best. But real breakthroughs emerge where there are (still) no reliable comparison values.
An example: If Airbnb had relied exclusively on historical market data from the hotel industry, the business model would never have emerged [5]. It didn’t fit into any existing market logic — neither in pricing structure, booking behavior, nor user expectations. The decision to continue developing the product was driven by user observation, cultural sensitivity, and the courage to prioritize intuition over numbers.
Numbers primarily deliver average values, but innovation thrives on outliers. The most exciting market opportunities are often still small, unclear, and contradictory. Those who respond to early trends, weak signals, and radical customer needs require courage for the incomplete. Intuition is not a mystical gut impulse [6], but rather an unconscious synthesis of experience, contextual knowledge, and systems understanding. It recognizes patterns before they manifest statistically — a decisive advantage in highly dynamic environments.
In emerging markets, user behavior, infrastructures, and value systems are in flux. Data from the past is therefore often misleading. Early-stage deeptech startups operate in complex, adaptive systems rather than linear cause-and-effect chains. Systems thinking helps recognize connections beyond the numbers: Where are power dynamics shifting? Which resources are becoming scarce? Which social movements are gaining momentum?
Data can only capture these dynamics with difficulty. Intuition, however, can anticipate them when combined with reflection, dialogue, and diversity in the team. For founders especially, it thus becomes a strategic capability: not as a replacement for data, but as a complementary compass in uncertain terrain.
Too Early for Metrics
In the early founding phase, there’s often a paradoxical relationship with data: On one hand, reliable metrics are lacking; on the other hand, founders are pressured to deliver exactly these for investor decks, accelerator programs, or internal progress tracking. The result: metrics are collected too early, misinterpreted, or simply overvalued. In the seed phase, however, it’s not about scaling but about understanding problems, relevance, and fit — aspects that can rarely be captured in dashboards [7].
A typical mistake: founders track conversion or retention KPIs before it’s even clear what exactly converts or why users stay (or don’t). Tools like Mixpanel or Google Analytics deliver numbers but no meaning. A 30% retention rate may look good on the surface but says nothing about whether the product actually solves a relevant problem. In this phase, it’s not the numbers themselves that count, but the stories behind them.
What really advances early-stage startups is not the hunt for data points, but the quality of user interactions. Those who directly observe how people interact with the product [8], who openly ask what they’re missing and why they’re interested in the first place, gain deeper insights than any funnel dashboard. Intuition plays a central role here: it helps distinguish between relevant and irrelevant user feedback and recognize implicit needs.
In the discovery phase, it’s about testing hypotheses, questioning assumptions, and understanding the “why” behind behavior. Data can support this exploratory work but cannot replace it [9]. Those who focus too heavily on quantifiable metrics in the seed phase are optimizing a model that may not even be viable yet. Instead: better five deep user interviews than five charts with superficial numbers.
In this early phase, intuition is not a risk but a necessity. It helps founders form patterns from signals, derive hypotheses from contradictions, and above all: sharpen their own judgment. Intuition doesn’t say “trust your gut” but rather “pay attention to what you hear between the lines.” Especially when the numbers are still silent.
Intuition Is Not Irrational
When founders speak about intuition, it’s often misunderstood: as diffuse inspiration, as the counterpart to “hard” data. But the opposite is true — at least when it comes to educated intuition [10]. Gerd Gigerenzer, Director at the Max Planck Institute for Human Development and one of the best-known intuition researchers, describes intuition as an “intelligent shortcut.” It’s not based on magic, but on experience, contextual knowledge, and implicit learning.
Gigerenzer describes intuition as part of human Fast-and-Frugal Heuristics: cognitive tools with which our brain makes good decisions even under uncertainty. Particularly in highly dynamic contexts like early-stage innovation, where data availability and market structure are still unclear, intuition delivers not only fast but often superior results. The reason: our brain recognizes patterns before they can be fully quantified.
What is described as a “gut feeling” is the sum of thousands of micro-experiences stored in the subconscious. In a founder context, this means: those who have conducted many customer interviews, iterated products, observed markets, or engaged intensively with a topic develop a fine antenna for what is (still) not visible but decisive. This “inner map” cannot be replaced by a dashboard.
The modern innovation environment is heavily focused on measurability and rationality. This has advantages, but also side effects. Intuition is devalued or neglected, even though it is a complementary resource. In a world that is increasingly complex and non-linear, we don’t need either-or logic, but the intelligent interplay of external data sources and internal experiential knowledge. Intuition is an evolutionary advantage.
While data is primarily descriptive (“What happened?”), intuition is often generative (“What could emerge?”). It helps form hypotheses, recognize blind spots, and ask bold questions. Founders who give their intuition space train their judgment ability and build a strategic early warning system that no KPI set in the world can deliver.
When Uncertainty Calls for Intuition
Deeptech founders work in an environment that is not complicated but complex. The difference is crucial: Complicated systems can be controlled with enough data, computing power, and expertise. Complex systems, however, are dynamic, unpredictable, non-linear. Here, precise predictions don’t help, only intelligent navigation. And this is exactly where intuition beats any Excel spreadsheet.
Some of the most important decisions in startups arise not from solid data foundations, but from inner resistance. The product doesn’t fit, even though the numbers look “okay.” The retention is right, but something is missing. Or: the market reacts differently than expected — not with disinterest, but with a completely different usage idea. Such signals rarely appear in metrics, but in conversations, side remarks, bodily reactions.
Successful pivots — like Slack (originally a game studio) [11] or Twitter (originally a podcast service) — didn’t begin with a KPI drop, but with a feeling: There’s something else here that generates more resonance. The numbers followed later.
In situations of high uncertainty, too much data can even be counterproductive. A study by Gigerenzer and Brighton [12] shows that in complex decision situations, less information often leads to better decisions — a phenomenon they call the “less-is-more effect.” Intuition unconsciously filters what is truly relevant, while data-driven analysis often gets lost in apparent objectivity.
Many startups measure what is measurable — not what would be meaningful. The result: masses of data that don’t help because the underlying questions are unclear or too unspecific. Instead of first formulating hypotheses that contribute to a relevant strategic question, data is often collected blindly. This is not only inefficient; it also obscures the view of what’s essential.
Intuition doesn’t come after analysis here, but before it. It helps recognize which questions are actually decisive — and which KPIs really say something about progress. Because not everything that can be measured is also decisive for product-market fit or user behavior in early phases. Those who think first and then measure use data as a tool. Those who measure first and then think follow the illusion of objective control.
In uncertainty, maximizing information density doesn’t help, but trusting the inner compass does. Those who wait too long for “more data” lose valuable time. Those who act too early, without context, risk activism. The art lies in using intuition as an early warning system and then validating it with targeted data points. This creates real decision quality: adaptive, courageous, and open to the future.
When Numbers Deceive and the Body Warns
The numbers look good: retention stable, CAC under control, first revenues flowing — but something feels wrong. The team is uncertain, the market reacts hesitantly, conversations fall flat. What seems analytically right doesn’t feel emotionally correct. This is exactly where intuition begins to take effect as an internal early warning system that recognizes patterns before they appear in data.
Neurobiological studies prove that the human body reacts to risks before they are consciously perceived. Neuroscientist Antonio Damasio speaks of somatic markers [13] in this context. These are physical sensations that work like an internal navigation system and help make better decisions in complex, uncertain situations. Founders who listen to their “gut feeling” often react faster to subtle signals. For instance, when a market doesn’t function as expected, even though all KPIs are green.
Intuition is not a magical flash of inspiration, but the result of years of implicit learning experiences. It connects sensory impressions, emotional reactions, and unconscious cognition into a decision impulse that often manifests as “it doesn’t fit” or “something’s not right here.” In the startup world, where founders operate with incomplete information, changing markets, and emergent technologies, this form of pattern recognition is particularly a competitive advantage in the early phase.
Confirming Rather Than Exploring: Where Data Really Shines
Not every decision in startup life takes place in the fog of uncertainty. In certain contexts, data is not just helpful but essential. For instance, when it comes to optimizing existing business models, scaling proven offerings, or developing incremental innovations. In these cases, clear patterns, validated assumptions, and comparable benchmarks already exist — the perfect playground for data-driven decisions.
Data doesn’t provide surprising insights here, but rather acts as a confirmation tool: it helps verify hypotheses, identify deviations, and fine-tune strategies. In established markets — such as expanding an e-commerce business, optimizing conversion funnels, or performance marketing — historical data provides reliable guidance. It reduces uncertainty because it’s based on a stable, repeatable environment.
A classic example: A/B tests. When testing two landing pages that differ only in their call-to-action, data delivers a clear result. The context is narrowly defined, the metrics clear, the goal measurable. Here, data-driven optimization is efficient, comprehensible, and valid, and should not be replaced by subjective assessments.
Even with incremental innovations — i.e., developments of existing products or services — data helps make well-founded decisions. An example is the use of analytics tools in product development: these enable detailed analysis of user behavior and thus improvement of existing functions. However, this only works when the usage context is already clearly defined, meaning when you know where you’re measuring and why.
Management thought leader Roger Martin [14] distinguishes between validated systems and new challenges: In validated systems, he writes, data are tools for refinement. In new challenges, however, creative model building is needed first — thinking in possibilities, not probabilities.
Don’t Measure Everything. Think First. Then Test.
In the early phase of a startup or when developing new ideas, a paradoxical behavior often emerges: masses of data are collected — without knowing what for. Analytics dashboards run hot, tracking tools collect clicks, dwell time, and bounce rates. And yet no real decision basis is created. Why? Because data without hypothesis has no direction.
The difference between blind analytics and real insights lies in the thinking beforehand: What do I actually want to know? What is my assumption and how would I refute it? Those who answer these questions cleanly can gain maximum clarity with minimal data use. Because data is not a substitute for thinking; it’s the tool that makes thinking testable.
The principle is not new. It’s at the core of Eric Ries’s Lean Startup approach [15]. In his framework, the “Build-Measure-Learn” cycle is central. But the crucial part comes before building: the hypothesis. What should be learned through the experiment? And which metric indicates whether the assumption is correct?
A good example: Instead of randomly developing new features and seeing how they perform, a startup first tests the hypothesis: “Our target group is willing to pay for additional personalization.” From this, a lean MVP can be built — perhaps a dummy feature behind a paywall — and measured against concrete conversions to see if the hypothesis is viable. No guesswork. No number salad. Just insight.
This approach is called hypothesis-driven experimentation and differs radically from what many startups call “data-driven”: There, numbers are often collected because it’s possible, not because one knows what to look for. The result: getting lost in meaningless KPIs, optimizing for irrelevant goals, and making decisions that support more than they lead.
Hypothesis-driven work forces a perspective shift: it’s not about proving an idea, but about refuting it. Only when an idea withstands well-designed tests does it gain substance. This mindset corresponds to the scientific method and protects against the dangerous tendency to seek data that confirms one’s own opinion. Keyword: Confirmation Bias (see above).
Data can be powerful, but only when it answers precise questions. Those who internalize this principle recognize: it’s not the dataset that’s decisive, but the clarity of the hypothesis. And sometimes that means measuring less — but measuring the right things.
When Gut Feeling Doesn’t Scale
However clear one’s own intuition may be in the founding routine — there are moments when a strong feeling alone isn’t enough. They weren’t part of the development process, know neither the informal user feedback nor the implicit market understanding. What for founders is a clear inner certainty remains initially invisible to outsiders. They want to see something. And not what “feels good,” but what can be calculated or measured.
Here begins the translation task: intuition must become quantifiable. Not because this makes it truer, but because others weren’t involved in the internal decision process. They need anchors. Reference points. Evidence. In such moments, data no longer serves primarily internal control, but external communication of credibility.
This becomes particularly clear when creating pitch decks: What often appears crystal clear to founders — the user need, the timing fit, the momentum — is initially just assertion to outsiders. Here KPIs help, even if they’re not yet perfect. Early indicators like engaged users, waitlist conversions, or time-to-first-value provide insights into traction before classic financial metrics take effect.
Data translates a feeling into a format that stakeholders can categorize. Just as good narratives convince emotionally, good data provides the rational foundation for trust. This balance is also described by VC fund First Round in its report on the characteristics of successful pitches [16].
This doesn’t mean everything must be measurable. But much is explainable if founders have the courage to translate their intuition into hypotheses and gradually underpin them with validatable data. And in this order: first feel, then structure, then communicate. Those who rely on numbers too early often lose sight of what’s really emerging. Those who start too late lose the trust of those who could help make it big.
Because investors don’t just invest in markets, but in convictions. But convictions without anchors seem like faith — and that’s not venture capital. If intuition is the starting signal for bold founding decisions, then data is the translation into a shared vision of the future.
Conclusion: Navigating Without a Map, But With a Compass
In a world where markets change faster than metrics stabilize, decisions without complete information are not the exception, but the rule. Founders don’t operate in the rearview mirror of data, but in the fog of uncertainty. Those who want to lead in this environment need more than analytical thinking — they need the courage to act even without guarantees.
Numbers are valuable, but they are never neutral. They measure what has been made measurable, not necessarily what matters. Intuition provides the context: it filters, structures, and first makes visible which data is actually relevant. Gerd Gigerenzer calls this “ecological rationality” [17]: situationally appropriate, experience-based action that in complex environments is often more accurate than mere calculation.
The decisive capability of modern founders lies not only in analyzing, but in deciding. That means: weighing risks without certainty. Testing hypotheses without prior proof. Intuition is not an esoteric gut feeling, but a condensed form of systems thinking, fed by experience, empathy, and pattern recognition.
Because exactly there — in the empty space between data points — the truly disruptive ideas emerge. Those that no investor demanded. Those for which there is (still) no market report. Those that address a problem that hasn’t yet been fully articulated. Trust your inner compass — especially when there’s nothing to measure.
Sources:
[1] https://www.cambridge.org/core/books/intelligence-of-intuition/E09D643BE8A3C1113E5F76BE1C4CA702
[2] https://www.penguin.co.uk/books/56314/thinking-fast-and-slow-by-kahneman-daniel/9780141033570
[3] https://www.seanellis.me/books#book-overview
[4] https://hbr.org/2019/09/dont-let-metrics-undermine-your-business
[5] https://customerculture.substack.com/p/how-airbnb-reached-product-market
[6] https://www.penguinrandomhouse.com/books/93618/the-power-of-intuition-by-gary-klein/
[7] https://www.ycombinator.com/library/4D-yc-s-essential-startup-advice
[8] https://www.ideou.com/blogs/inspiration/what-is-human-centered-design
[9] https://steveblank.com/2009/12/17/building-a-company-with-customer-data-metrics-are-not-enough/
[10] https://www.penguinrandomhouse.com/books/309675/risk-savvy-by-gerd-gigerenzer/
[11] https://review.firstround.com/podcast/lessons-from-slack-on-decision-making-product-led-growth-and-taking-big-swings-noah-desai-weiss/
[12] https://onlinelibrary.wiley.com/doi/full/10.1111/j.1756-8765.2008.01006.x
[13] https://www.penguinrandomhouse.com/books/297609/descartes-error-by-antonio-damasio/
[14] https://hbr.org/books/playing-to-win
[15] https://leanstartup.co/books/
[16] https://review.firstround.com/the-30-best-pieces-of-advice-for-entrepreneurs-in-2023/
[17] https://u-paris.fr/diip/gerd-gigerenzer-from-bounded-rationality-to-ecological-rationality/
