Original source: The Jeff Bullas Show
This video from The Jeff Bullas Show covered a lot of ground. 9 segments stood out as worth your time. Everything below links directly to the timestamp in the original video.
What happens when a marketing executive decides that 'trust us, it works' is no longer an acceptable answer — and goes back to mathematics to prove it? The result is a company, and a career-defining obsession.
Proof Analytics Born From a Marketing Executive's Refusal to Accept 'Unproven' as an Answer
The origin of Proof Analytics is not a tidy founder story — it begins with wounded professional pride. Working at HP, Stouse grew increasingly frustrated when business leaders dismissed marketing as a cost centre that couldn't demonstrate measurable return. His response was not to argue back but to go further: pulling out old mathematics textbooks, studying the econometric methods Procter & Gamble had pioneered, and finding an unexpected ally in HP CFO Bob Wayman. By 2008 and 2009, Stouse had built a reputation for constructing causal analytics portraits — cause-and-effect maps connecting marketing activity to business outcomes — compelling enough that executives began calibrating decisions against them. The gap that remained was scale: the work was manual, bespoke, and impossible to operationalise across organisations. That gap became the product.
What most people miss in this kind of origin story is the structural tension it exposes. Marketing has historically occupied an awkward position in the corporate hierarchy — indispensable in theory, discretionary in practice. The real question is not whether marketing works but whether it can be made legible to a CFO's model of the world. Stouse's journey from ego-driven frustration to software founder is, at its core, an attempt to answer that question with mathematics rather than persuasion.
"From a purely ego point of view, that really pissed me off."
An HP CEO's Physical Confrontation Became the Catalyst for a New Analytics Company
The moment that most directly accelerated Stouse's analytical reckoning came during a meeting with HP CEO Mark Hurd. Hurd, dissatisfied with marketing's inability to justify its spending, physically backed Stouse against a wall — close enough that the distance between them was measured in inches rather than feet. The implicit message was unambiguous: produce evidence or prepare to leave. What followed was not resentment but motivation. Stouse treated the encounter as a forcing function, channelling the pressure into the analytical work that would eventually become Proof Analytics. Hurd, before his death, reportedly took credit for the breakthrough that followed — a detail Stouse recounts without bitterness.
The structural reality here is that extreme accountability pressure, applied at the right moment to someone with the capability and disposition to respond constructively, can generate genuinely productive outcomes. If the pressure had landed on someone without Stouse's mathematical curiosity, it would simply have produced a resignation letter. What transformed the confrontation into a founding story was not the pressure itself but the direction in which it was channelled — outward into problem-solving rather than inward into defensiveness.
"He backs me up against the wall — inches from my nose. And that's a moment you don't forget."
Most Corporations Don't Have Enough Data to Benefit From AI, Proof Analytics Founder Argues
Proof Analytics was built as an AI-first product from its founding in 2015, but Stouse draws a sharp distinction between the AI embedded in his platform and the machine-learning applications that dominate mainstream discussion. Machine learning, he argues, is fundamentally pattern-detection at scale — useful when enormous datasets exist, but structurally unsuited to causal analytics, where the goal is to explain why outcomes occur rather than merely identify what patterns accompany them. More pointedly, Stouse contends that most corporations simply do not possess the volume of data that meaningful AI applications require. The genuine big-data environments — pharmaceutical research, aerospace, automotive — are sectors where life-or-death stakes drive data accumulation. Large technology companies, counterintuitively, often fall short of that threshold for their internal analytics purposes.
The implication is consequential for any organisation currently building an AI strategy around the assumption that more data automatically enables better decisions. The structural reality is that AI's analytical leverage depends entirely on data volume and causal clarity — and confusing pattern recognition with causal understanding is precisely the kind of error that produces confident wrong answers. Stouse's framing suggests that for the majority of enterprises, the more rigorous path runs through econometric regression rather than machine learning.
"Most corporations don't have a lot of big data. They have a lot of data, a lot of time series data — but not big data."
Regression Analysis Is a GPS for Business Decisions, Not a Rearview Mirror
The most common misreading of regression analysis, according to Stouse, is to treat it as a sophisticated form of extrapolation — projecting a past trend forward in a straight line. The distinction matters enormously. Extrapolation ignores externalities and assumes the future will mechanically resemble the recent past. Regression, by contrast, identifies the structural relationships between multiple variables across historical data — accounting for seasonality, competitive dynamics, and other contextual factors — and uses those relationships to generate forecasts that are recalibrated continuously against incoming actuals. Proof Analytics rebuilds its models daily, comparing forecasts to real-world results and surfacing the gap when divergence appears. The operational analogy is a GPS recalculating a route in response to traffic: not simply predicting the original arrival time but actively offering an alternative path to get back on schedule.
The GPS metaphor is more than rhetorical convenience — it reframes what analytical tools are actually for. The real question is not 'what happened?' but 'given where we are now, what is the fastest route to where we need to be?' That navigation framing shifts analytics from a reporting function to a decision-support function, which is a meaningful architectural distinction for any organisation still using dashboards primarily to explain the past.
"The future becomes the present, it gets checked, and everybody goes — wow, we're tracking almost exactly. Or: there's a gap between forecast and actuals. What's causing that?"
Analytics Products Can't Ship as MVPs — Proof Analytics Took Three Years and 19 Paying Customers to Build
The conventional startup playbook prescribes a minimum viable product released quickly to test market assumptions. Stouse rejected that approach for a specific structural reason: analytical products, unlike most software, are binary in their utility. An incomplete analytical model doesn't merely lack features — it produces wrong answers, which is categorically worse than no answers at all. The solution was unconventional. Proof Analytics spent three years building the product with roughly 18 to 19 early customers who paid not in cash but in detailed, continuous feedback. Those customers, because they were pre-release participants rather than paying clients with service expectations, could engage with the product at a level of granularity the open market cannot sustain — dissecting individual components, tolerating instability, and providing the kind of atomised critique that iteration requires.
The deeper lesson Stouse draws from that period is less about product strategy than about psychological tolerance. Entrepreneurial development, in his account, is structurally a condition of moving through dense fog with incomplete information — aware that much of what felt like sound judgement a year ago was actually ignorance operating with confidence. The structural reality is that accepting that condition, rather than fighting it, is the minimum requirement for staying in the game long enough to learn enough to matter.
"You feel like you're dodging bullets in a fog of ignorance. The temptation is to say to yourself — I have no idea what I'm doing."
A Retailer's Hourly Data Feed Revealed That Every Business Decision Is a Navigation Problem
The product insight that unlocked Proof Analytics' interface came not from a design sprint but from a specific customer behaviour. A large retailer was feeding data into the platform hourly rather than at the daily intervals the system was designed around. The effect was visible and unexpected: the analytical models began recomputing in near-real time, updating their outputs continuously as new data arrived. Watching it happen, Stouse and his team recognised the behaviour immediately — it was identical to watching a GPS recalculate a route. The metaphor that had previously been a loose analogy suddenly became precise product direction. The broader principle that crystallised from that moment was the claim that all business questions are, at their core, navigation questions: where are we, where do we need to go, and what is changing around us that affects the path between the two.
What makes the GPS metaphor analytically useful rather than merely decorative is its emphasis on continuous adjustment. Navigation is not a one-time calculation — it is a feedback loop that incorporates new information and updates the recommended path in response. If that is the correct model for business decision-making, then the dominant approach — quarterly reviews of historical data followed by annual planning cycles — is not just slow. It is architecturally misaligned with the nature of the problem it is trying to solve.
"It was just like a GPS — literally like watching the GPS on your phone recalculate. And we were like: oh my God, that's it."
The Feature That Built Trust in GPS Navigation Also Unlocked Confidence in Marketing Analytics
When Proof Analytics introduced its GPS-style interface, the feature that generated the most significant shift in user confidence was not the forecasting itself but the visible, interactive countdown to predicted outcomes — a design parallel to the estimated time of arrival display that GPS providers introduced around 2012 and 2013. Stouse observed that the ETA countdown on navigation apps was the single feature most responsible for users trusting the technology, because it made a specific, falsifiable prediction and then delivered on it — arriving at a restaurant at precisely the moment the screen reached zero. Proof replicated that dynamic by allowing users to watch forecast accuracy resolve in real time, and to war-game alternative investment scenarios on the fly: redirecting $10 million from one channel to another and observing the projected outcome update immediately.
The structural insight is that confidence in analytical tools is not primarily a function of mathematical rigour — it is a function of legible accuracy. Users do not evaluate regression models by inspecting their coefficients. They evaluate them by watching predictions come true. The implication for any organisation deploying analytics is that the gap between a technically sound model and one that actually changes behaviour often has less to do with the mathematics than with whether the interface makes accuracy visible in real time.
"Nothing made a bigger difference in people's confidence in the GPS than that ETA countdown — because when you pull up to the restaurant and it goes to zero, you go: okay, this machine knows its stuff."
Stouse on Entrepreneurship: 'Suffering' Is the Right Word, and Most People Don't Understand Why
Stouse uses the word 'suffering' deliberately when describing the entrepreneurial experience, and the precision of the choice matters. The cultural narrative around founders and CEOs — lionised publicly, assumed to be insulated by wealth and status — actively compounds the difficulty by making the isolation harder to name or discuss. Stouse argues that the loneliness particular to executive leadership is structurally unlike ordinary professional hardship, and that people who have not experienced it are poorly positioned to evaluate it from the outside. His prescription is not resilience in the motivational-poster sense but something more specific: cultivating a form of joy that is independent of circumstances, capable of coexisting with genuine difficulty rather than requiring its absence. He frames this as expanding one's perceptual aperture — holding what exists alongside what is possible, rather than collapsing attention onto what is going wrong.
What sharpens this beyond standard founder reflection is the data-scientist framing Stouse brings to the question of control. Outcomes, he argues, are always the product of variables both within and beyond an individual's influence — and a significant portion of what determines success or failure will never be fully understood, let alone managed. The discipline is not to pretend otherwise but to act with full commitment while holding the structural reality that the outcome function contains terms you will never get to see.
"Entrepreneurship has some really significant suffering. I choose that word on purpose — it's a lot harder than it looks."
Da Vinci's Hidden Co-Signature Offers a Closing Argument for Collaborative Creation
Stouse closes with a historical detail that functions as more than decoration. In private diary entries, Leonardo da Vinci sketched designs alongside a hidden joint signature — his own name and that of Lorenzo de' Medici, encoded together. Da Vinci, by Stouse's account, was quietly giving his patron equal creative billing: recognising that Medici's resources, vision, and patronage were not merely logistical support but genuine co-authorship of what the work became. The anecdote reframes individual genius as collaborative architecture — a structure in which the person whose name history remembers is, in their own private reckoning, only one of the necessary conditions for the thing to exist at all.
The real question the story poses is whether that kind of private acknowledgement reflects intellectual honesty or something more fundamental: a recognition that creation, at the level where it actually matters, is never singular. For someone who built a company by synthesising econometric methods from Procter & Gamble, mathematical frameworks from academic literature, and operational insight from customers who paid with feedback rather than money, the Medici parallel is less metaphor than self-description. What most people miss is that the myth of the solitary founder, like the myth of the solitary genius, is almost always a compression of a much more distributed truth.
"Da Vinci is giving equal billing to Medici — recognising the co-creativity that Medici had brought to the table."
▶ Watch this segment — 1:02:12
Summarised from The Jeff Bullas Show · 1:04:53. All credit belongs to the original creators. Streamed.News summarises publicly available video content.
