Breakthrough AI Operators

Roland Siebelink
  • 0.0 (0)
  • Management
  • Updated Weekly

Breakthrough AI Operators is a podcast about how the best startup founders are reinventing how their companies work. Not AI hype. Not vendor pitches. Real operators who've rebuilt significant parts of their business around AI — and can talk honestly about what worked and what didn't. Hosted by Roland Siebelink and Doug Miller, co-founders of Midstage Accelerator (7 unicorns built between them, 100+ leadership teams scaled), each episode features a founder who's achieved a genuine step-change breakthrough in how their company operates. These aren't productivity wins or tool adoption stories — they're companies that are structurally different because of AI. If you're a founder at a 20–300 person company actively figuring out what AI means for your operating model and competitive position, this show gives you real stories from people in the field — not consultants theorizing from the sidelines.

  1. Aug 6

    Don't Think AI. Think Business First. | Iveda | David Ly | EP 209

    Picture a city government worried about crime: robberies, car accidents, crooks on the street. An AI vendor walks in pitching license plate recognition and predictive policing. It sounds right. Except the city's more urgent problem might be flooded streets, and no security vendor ever thought to ask about that. One AI CEO built his entire company around asking that question first, which is the opposite of how most AI startups go to market. David Ly founded Iveda in 2003 selling IP camera monitoring as a service, built one of America's first human video monitoring centers with 60 people watching screens around the clock, then shut it down once he realized humans couldn't do the job well and trained machine learning on the footage they'd collected instead. He took the company public on NASDAQ in 2022, not to cash out, but to acquire more. Ly's rule for founders is blunt: don't think about AI first, get to know your business and your real, unflattering problems, and only then ask where AI helps. He walks through a real example: a municipality that Iveda assumed cared about crime prevention actually cared more about flooded streets, so his team built detection for that instead of pitching the license plate features they'd planned to sell. They brought the fix back to that same customer before ever marketing it. He calls most competitors "not wrong, just not accurate," because they pitch a bundle of goods before they've confirmed what's actually broken. He also still flies out to shake a customer's hand personally, even running an AI-native business, because that relationship is what makes a deal stick. The bigger turn in the conversation is personal. A few years ago, Ly's newly hired chief of staff asked him a question that stung: if you're this good and this smart, why are you still running a six or seven million dollar company instead of fifty? That forced him to rebuild Iveda's hiring and culture from the ground up, prioritizing attitude and total alignment across every geography the company operates in over pedigree or credentials, and building a company-wide habit of admitting what you don't understand instead of nodding along in meetings full of acronyms. Roland sees the flooding example as the sharpest version of a mistake he watches SaaS founders at the $1M to $50M stage make constantly: reaching for the AI feature that sounds impressive before confirming what the customer is actually stuck on. The other pattern worth sitting with is how much Ly credits one blunt question from someone close to him for a breakthrough his own instincts couldn't produce. Key Moments: 02:37 — Why "we are a solutions provider" is a cliché, and what actually breaks AI vendors in the field 04:27 — The flooding example: pitching crime prevention when the real problem was something nobody had asked about 07:14 — Why "technology makes anybody look and sound clever," and why that's never enough 09:07 — Still flying out to shake a customer's hand personally, even in an AI-native business 09:49 — Buyer persona versus user persona, and why users just want minimal thinking 13:59 — Getting full market intelligence through a reseller channel network instead of a direct sales team 20:53 — "Why are you still sitting here running a six, seven, ten million dollar company?" 23:01 — Rebuilding hiring around attitude and total alignment instead of degrees and pedigree If the question that changed everything for Ly is one you haven't been asked yet, Roland works through exactly that kind of stuck point with SaaS founders inside Midstage. [mdstg.ac/drag-erase] #AIVideoSurveillance #SmartCityAI #AIVideoAnalytics #FounderGrowth #BreakthroughAIOperators

    Don't Think AI. Think Business First. | Iveda | David Ly | EP 209

  2. Jul 29

    Product Ahead, Then GTM Ahead | SpotDraft | Shashank Bijapur | EP 208

    What do you do when the exact data your AI needs to get smart is legally required to stay private forever? A confidentiality agreement doesn't show up in a public training set, by definition it can't. One founder discovered that gap years before generative AI existed, and had to build an entire operating platform just to earn the right to the data his own product needed. Shashank Bijapur is co-founder and CEO of SpotDraft, an AI-native contract lifecycle management platform he built with his co-founder Madhav after leaving a law career at White & Case. SpotDraft has processed more than a million contracts, raised a $54 million Series B, and recently became the first CLM to run contract analysis entirely on-device. The idea started on a New Year's Eve in 2017, when Bijapur was running due diligence reports instead of celebrating, read a headline about Elon Musk building self-driving cars, and couldn't square that cars were driving themselves while he was still copy-pasting contract language by hand. He and Madhav's first hypothesis was straightforward: train a model on legal contracts and automate redlining. It fell apart fast, because confidentiality agreements are confidential. There was no real training data to learn from, only invented data that made accuracy a coin flip. So they took two steps back and built the actual infrastructure lawyers were missing, workflows, a repository, a real system of record, which is what eventually became the CLM. AI was never bolted on afterward. It's the reason the company exists. That same instinct shaped a more recent bet. Industries like pharma and defense kept raising the same objection: where does our data go once it touches an LLM? SpotDraft partnered with Qualcomm to run contract analysis entirely on-device, in airplane mode, with nothing leaving the machine, making it the first CLM to offer that option. Bijapur also described becoming his own company's bottleneck around year five, answering every question himself until his team stopped trying to decide anything without him, a loop his COO broke by introducing OKRs so people knew exactly what they could decide without asking. Roland has seen the founder-as-bottleneck pattern often with SaaS leaders at the $1M to $50M stage, and what he found most useful here is how directly Bijapur named the fix: it's not about hiring smarter people, it's about explicitly telling them what they're allowed to decide without you. The other pattern worth sitting with is SpotDraft spending three full years in pure product mode with no go-to-market function at all, on the belief that great technology speaks for itself. It doesn't, and Bijapur said as much. Key Moments 01:49 — The origin story: a recovering lawyer, a due diligence report, and a New Year's Eve at White & Case 02:18 — Reading about Elon Musk's self-driving cars while still copy-pasting contract language by hand 04:05 — Why training an AI model on legal contracts almost couldn't work: confidential means confidential 10:42 — "After Shakespeare, what's the most difficult language you can read? That damn contract" 12:27 — The Qualcomm partnership, and why on-device AI mattered for pharma and defense clients 18:16 — Realizing he had become his own company's bottleneck, and what broke the loop 20:22 — Getting product and go-to-market to finally run in sync instead of chasing each other — If you recognize the moment where your own team stopped deciding anything without checking with you first, that's exactly the kind of stuck point Roland works through with SaaS founders inside Midstage. mdstg.ac/drag-erase #AIContractManagement #LegalAI #CLMPlatform #FounderGrowth #BreakthroughAIOperators

    Product Ahead, Then GTM Ahead | SpotDraft | Shashank Bijapur | EP 208

  3. Jul 14

    20 Years to Build the Product People Cry Over | Mixbook | Andrew Laffoon | EP 207

    Andrew Laffoon is co-founder and CEO of Mixbook, a photo book platform he has led for 20 years—from hobbyist niche to mainstream storytelling product. In this episode, Andrew explains how Mixbook built Story Mode (an AI-powered photo book creator powered by OpenAI, Claude, and Google, launched 8 days before this recording), what stalled growth after the first $25M, and why the exec team is never the answer when the bottleneck is the founder. Topics covered: - Why 98% of people couldn't make a meaningful photo book before Story Mode—and what that says about storytelling as a market - The exec-team-as-mentors trap: why hiring smart people can't fix a founder who hasn't grown into the role yet - Flywheel thinking: how to identify the bottleneck, measure it, and compound growth on a schedule - "The first job of a CEO is to define reality": what this means for a company at 130 people - Sam Altman's persuasion warning—and what a founder deploying AI should actually be worried about --- Mixbook is offering listeners 50% off their first order with code BREAKTHROUGH. Worth trying if you want to see what Story Mode does with your own photos before deciding whether AI-written narration actually holds up. https://bit.ly/4fvDI64 If you recognize the wall Laffoon described, growth that's stalled for reasons you can't quite name, Roland works through exactly that kind of stuck point with SaaS founders inside Midstage. mdstg.ac/drag-erase #AIPhotoBook #StoryDrivenAI #PersonalizedPhotoGifts #FounderGrowth #BreakthroughAIOperators

    20 Years to Build the Product People Cry Over | Mixbook | Andrew Laffoon | EP 207

  4. Jul 7

    300M Calls, 70 People: How Vertical AI Beats Horizontal Every Time | Flip | Brian Schiff | EP 206

    Would you turn away a paying customer to protect a product that isn't ready for them? Last year, a major telecom company came to a fast-growing AI startup wanting to buy in. The founder said no, not because the money was bad, but because his product hadn't earned the right to serve that industry yet. Brian Schiff is co-founder and CEO of Flip, the vertical voice AI platform that grew out of a failed idea about college ride-hailing. Schiff and his co-founder Sam met at Cornell and built a ride-hailing app for taxi companies in upstate New York, where ridesharing itself was still technically banned. The app worked fine, but the phones at these taxi companies never stopped ringing. That was 2018, right as Alexa and Google Home had their moment, and it pushed the pair toward what Schiff calls a product pivot rather than a market pivot: they kept selling to the same customers, just sold them something new. Cornell's eLab program had drilled one lesson into them, be concrete about the problem and who you're solving it for, and that discipline shaped everything after. It also shaped Flip's contrarian bets. Most AI customer service companies build the technology first and go looking for somewhere to apply it, what Schiff calls technology outward instead of customer inward. Flip built industry by industry instead, transportation first, then retail, then healthcare, turning down the telecom client because an untrained agent would have been, in his words, the dumbest agent you've ever talked to. Flip also priced itself on usage from day one, charging only for resolved calls, a model investors compared to payment processors, which get valued at a fraction of typical SaaS multiples. Schiff built it anyway because it removed the buyer's risk. Roland has coached plenty of SaaS founders who feel the pull to say yes the moment a bigger check shows up, and Schiff's telecom refusal is the clearest version of that discipline he's heard on the show. What stands out isn't the refusal itself, it's that Flip proved out transportation completely before touching retail, and retail before touching healthcare. That sequencing is the exact pattern Roland wishes more founders at the $5M to $15M stage would follow instead of chasing logo diversity to look bigger than they are. Key Moments: 02:00 — The "ten years to overnight success" story, and why it started in taxi dispatch instead of AI 05:40 — A market pivot versus a product pivot, and why Flip chose the harder one 07:42 — Technology outward versus customer inward: the split Schiff says defines the AI customer service race 11:05 — The usage-based pricing bet investors warned would cap Flip's valuation 18:23 — Turning away a telecom client because the agent "would not be able to do anything" for them 24:19 — Building a client reference list as long as the client list itself 27:53 — "Everything in this company is written in pencil, not pen" If turning down a revenue-adjacent client to stay disciplined about who you actually serve is a decision you're wrestling with, Roland works through exactly that tradeoff with SaaS leadership teams inside Midstage's advisory work. mdstg.ac/drag-erase #VerticalAI #VoiceAI #AICustomerService #GoToMarket #BreakthroughAIOperators

    300M Calls, 70 People: How Vertical AI Beats Horizontal Every Time | Flip | Brian Schiff | EP 206

  5. Jun 30

    From 50 to 5,000 Customers in a Month: How Pictory Clawed Its Way to Product-Market Fit | Pictory | Vikram Chalana | EP 205

    In 2019, before ChatGPT, before Sora, before most people had heard the word "foundation model" used in a business context, one founder made a call that looked strange at the time: he decided not to train his own AI models. Not because he couldn't. Because he believed the models would arrive — and that the real value would be in the workflow built on top of them. Vikram Chalana took his first company, Winshuttle, from zero to 350 employees and 2,000 customers across 66 countries before selling it to Symphony Technology Group in 2018. He's now co-founder and CEO of Pictory, an AI video creation platform that has brought 20,000 organisations away from traditional video production with a team of 60 and total funding of $4.7M, in a market where direct competitors have raised hundreds of millions. The primary thread of this conversation is capital efficiency as deliberate strategy — not constraint. Vikram describes the founding decision to compose best-in-class AI infrastructure, integrating ElevenLabs, Veo, HeyGen, and others rather than building proprietary models, as the single biggest factor in Pictory's margins and speed. The workflow layer is the moat. The model underneath it is swappable. That logic, which felt contrarian in 2019, has played out in ways that are now visible in the market: Pictory is profitable; some of its better-funded competitors are still explaining what they spent the money on. The second thread is how product-market fit actually arrives — and what it looks like when it does. In summer 2021, Pictory went from 50 paying customers to 5,000 in a single month, after running an AppSumo campaign that found a completely different customer segment than the enterprise marketing teams the company had been chasing. The insight Vikram draws from that: product-market fit is a combination of product and market, and the same product on a different market is a different company. Before that AppSumo experiment, Pictory was struggling. The product has not changed, the customer has. Roland notes that Vikram's approach to the AI model question is a sharper version of what he sees in the most capital-efficient SaaS companies he works with — founders who identified one specific structural bet early and stayed disciplined about it, not because they had certainty, but because the logic of the bet remained sound as new information arrived. The founders who raised heavily to build proprietary AI infrastructure are now carrying that investment on their balance sheets. The founders who stayed composable are building on top of the same underlying models, at lower cost, with more flexibility. Vikram's willingness to name that dynamic directly is unusual — and useful. Key Moments: 00:00 — From 50 to 5,000 customers in a month: what product-market fit actually feels like from inside it 02:46 — Why Pictory decided not to train its own AI models in 2019, and what that bet looked like before anyone knew it would pay off 08:57 — The current AI model stack: how Pictory integrates Claude, OpenAI, Google Veo, ElevenLabs, HeyGen, and others, and what swappability actually buys you 12:11 — Why workflow stickiness matters more than model quality: the front-end moat Pictory is betting on 14:05 — The AppSumo experiment: how switching markets, not products, took Pictory from struggle to 5,000 customers in four weeks 18:14 — CEO vs. CTO: how Vikram draws the line between the two roles, and what the Winshuttle transition taught him about giving up control 22:24 — Working on the business, not in it: what the phrase actually means, and why knowing it doesn't make it any easier to do 27:06 — The Pictory CREDO (Curiosity, Respect, Expeditiousness, Data, Openness), and what happens when a core value gets tested by a layoff or a firing 22:46 — Why the person who got you to $1M might not be the right person to get you to $10M, and why that's not a failure on their part 24:43 — Why good products sometimes die because the founder got bored: Vikram's discipline around not rebuilding what already works 29:47 — Vikram's one piece of advice for newer founders: stop being so risk averse, just jump in — Pictory offers a free trial at pictory.ai — worth visiting if you're a founder or marketer who wants to see what AI video production actually looks like without a production budget. If you're navigating the AI stack question — whether to build, buy, or compose — and want to pressure-test the logic of your bet against what Midstage has seen across dozens of software companies at the $1M–$50M stage, that's a conversation worth having. mdstg.ac/drag-erase #AIVideoCreation #TextToVideoAI #AIVideoForMarketing #AIVideoGeneration #BreakthroughAIOperators

    From 50 to 5,000 Customers in a Month: How Pictory Clawed Its Way to Product-Market Fit | Pictory | Vikram Chalana | EP 205

  6. Jun 23

    Unicorn in the Making: The Playbook for Scaling from $1M to $100M | Midstage Accelerator | Payandeh Ekrami | EP 204

    Most founders believe their team isn't aligned because the team isn't capable enough. Almost none of them have considered that the team can't align around a direction that lives entirely inside the founder's head. That distinction — between a team development problem and a founder bottleneck problem — is the one that determines whether a scaling company breaks through or breaks down. Payandeh Ekrami is a seasoned enterprise technology sales executive with senior leadership roles at Snowflake, Anaplan, and Samsung, specialising in helping high-growth companies build and scale revenue organisations. In this episode, she turns the mic around on Roland — interviewing him about the patterns he's observed across three unicorn journeys and what those patterns mean for any founder navigating the mid-stage transition today. The conversation surfaces two things that rarely appear together in the same discussion. The first is Roland's diagnostic of why founder-led companies stall — not at the market or product level, but at the operating level, when the habits that built the business start blocking the next stage of growth. The second is the more personal thread Payandeh draws out: Roland's underlying conviction that every person can find a context where they genuinely excel, and that one of the most important things any leader can do — including a founder — is help the people around them find that fit rather than forcing them into shapes that don't suit them. Roland names several specific tools and moments that illustrate how Midstage Accelerator does its work. The first workshop exercise — asking every leadership team member to write down independently what they believe they're responsible for — has made teams cry from the frustration of seeing how misaligned they actually were, with a head of engineering optimising for zero outages while the CEO expected ten monthly product launches. The practice of asking a founder CEO to sit quietly and speak last in workshop discussions reliably reveals which team members have been contributing original thinking and which have been echoing the founder back. And the question Roland poses before every engagement — do you work for yourself, or do you work for the company? — functions as a direct filter for whether the founder is building a business or building a legacy. Roland notes that the founders who reach the mid-stage with the strongest track records are often the hardest to work with at first — precisely because their instinct to be the smartest, most decisive person in the room is what got them there. The move from managing by assignment to managing by area, from controlling every detail to setting outcomes and trusting people to own their domains, is not a knowledge problem. It is a psychological one. What changes the equation, in his experience, is not convincing founders to delegate — it is lighting a bigger fire for what becomes possible when they do. Key Moments: 00:00 — Roland's opening claim: why seeing the movie three times is the core value he brings to founders 02:16 — What makes the mid-stage different from early-stage startup and from running an established company 05:50 — Why Roland calls himself an accelerator, not a coach — and what the distinction actually means 08:24 — The workshop instruction Roland gives every new founder CEO: sit down, be quiet, speak last 10:02 — What that silence reveals about who's actually thinking on the team versus who's been echoing the founder 12:52 — The alignment exercise that has made leadership teams cry: write down what you think you're responsible for 14:19 — The one question Roland asks before signing any engagement — and why the answer is a green flag or a red flag 16:00 — The Orange Communications "today and tomorrow" model: how to keep innovation alive without burning down operations 20:00 — The flywheel framework: four functions, one simple picture, and why it makes delegation feel possible 25:04 — What Roland looks for before taking on a client: ambition, coachability, and pride in the craft of leadership If you're a founder CEO navigating the transition from doing everything yourself to leading through a team, and you want a diagnostic on where the real bottlenecks are, reach out to Midstage Institute to start the conversation. mdstg.ac/drag-erase #StartupScaling #UnicornFounder #FounderCEOCoaching #MidStageStartup #BreakthroughAIOperators

    Unicorn in the Making: The Playbook for Scaling from $1M to $100M | Midstage Accelerator | Payandeh Ekrami | EP 204

  7. Jun 16

    The Boring Part Is the Defensible Part | Evercam | Marco Herbst | EP 203

    What does it actually take to build an AI-powered construction intelligence platform — and why does it take 12 years? Most founders trying to layer AI onto physical industries discover too late that the data problem is the real problem. The model is only as good as what you fed it, and construction sites are the least controlled environments on earth. One company solved that problem by not trying to skip it. Marco Herbst is the founder and CEO of Evercam, a Reality Driven Intelligence platform that has accumulated 13,000 years of labeled construction video across 2,500+ customers including Intel, Microsoft, Meta, and Shell — built over 12 years of deliberate, unglamorous infrastructure work before the AI moment arrived. Marco's path to construction wasn't linear. He co-founded Jobs.ie in the early internet era and sold it in 2005, spent five years in Berlin by design, then returned to Ireland drawn by something specific: cameras as a communications tool, and the growing gap between what IP cameras could do and what construction sites actually needed. The early years weren't about intelligence — they were about reliability. Keeping a camera alive on a site for 18 months without maintenance, in weather, in dust, in a world where a late-detected problem adds zeros to the cost. That obsession, which looked like constraint, became the foundation for everything that followed. The second thread of the conversation is equally useful for any founder navigating the scale-up transition. Marco describes hiring three successive commercial leaders across the company's growth arc — each one suited to a different stage — and the structural decision to divide Evercam into regional P&L profit centers when telling a coherent story across eight global markets became impossible from the center. He also introduces the Evercam 100: a publicly available library of 100 construction workflows that deliver measurable ROI, hosted at workflows.evercam.io. Marco's reasoning for publishing it openly — that sharing information leads to better outcomes over the long arc — is both a genuine conviction and a strategic bet on collaboration winning over protection. Roland notes that Marco's story is the clearest example he's encountered of a founder earning the AI advantage the slow way — by doing the data work no one else was willing to do, long before AI made the data valuable. In his advisory work with SaaS companies at the $1M–$50M stage, Roland consistently sees founders who want to compete on intelligence before they've built a reliable data layer. The founders who resist that shortcut — who treat the infrastructure as the product until it actually works — tend to arrive at the AI moment with something competitors can't replicate quickly. Marco's 13,000 years of labeled construction footage is the most concrete version of that principle Roland has seen. Key Moments: 00:00 — Marco's definition of product-market-fit: the 2am call from Texas 02:05 — Why Evercam chose hardware reliability over AI — and why it was the right call 03:01 — From Berlin sabbatical to construction cameras: how Marco spotted a gap no software founder was looking at 05:22 — The moment construction clicked — and why security cameras never did 07:17 — How the mission-critical sector (Intel, Microsoft, Meta, Shell) took Evercam global 09:25 — Dividing into regional P&L profit centers to manage 8 global markets coherently 10:08 — Three commercial leaders across three growth stages: zero to one, one to ten, ten to one hundred 13:32 — How VLMs transformed what Evercam can detect — from counting vehicles to spotting unscheduled hazardous activity 16:43 — The Evercam 100: why they published 100 customer ROI workflows openly, and what it signals about their culture 19:52 — Why Marco published the workflows: a genuine conviction that collaboration wins over the long arc 21:32 — From "font police" to "management by joy": Marco's evolution as a founder and what it means in practice — If you're navigating the transition from founder-led construction sales to a scalable commercial model — or trying to understand how AI creates defensibility in physical industries — contact us to tell us about your project at https://www.evercam.com/roi-calculator. If scaling a hardware-software business across global markets is a challenge you're working through right now, Midstage works with founders at the $1M–$50M stage on exactly this kind of inflection. mdstg.ac/drag-erase #ConstructionTechnology #ConstructionIntelligence #AIConstruction #FounderLedSales #BreakthroughAIOperators

    The Boring Part Is the Defensible Part | Evercam | Marco Herbst | EP 203

  8. Jun 9

    Most Companies Are Measuring Marketing Wrong — And AI Won't Save Them | Hawke Media | Erik Huberman | EP 202

    Most founders have more marketing data than they know what to do with — and are making worse decisions because of it. The issue isn't measurement. It's that the metrics they're watching were never designed to match the sales cycle they're actually running. When you compare this week's ad spend to this week's revenue in a business where the average purchase takes a month to close, you will always misread what's working. Erik Huberman has watched this exact pattern destroy good campaigns — and has the data from 5,000+ brands to prove it. Erik is the founder and CEO of Hawke Media, a full-service outsourced marketing agency that has helped over 5,000 brands — including Red Bull, Verizon, and Crocs — generate nearly $3 billion in client revenue. He built Hawke from scratch into a 250-person bootstrapped company and spent eight years turning client data into a proprietary AI platform trained on over $700 million in media spend. The conversation opens on a problem Erik sees across every category: founders who look at ROAS on a weekly or monthly basis and draw conclusions that are structurally wrong. If a sales cycle averages 30 days, scaling ad spend from $1K to $5K daily won't show up in revenue for two months — but most founders see flat revenue, conclude the channel doesn't work, and cut the spend that was compounding. Erik's argument isn't that data is bad. It's that a correct data point, read through the wrong frame, produces confident, wrong decisions. He extends this into a broader claim about the AI hype cycle: that much of what AI is being credited with is the same problem — regurgitating inputs without understanding the context that makes data meaningful. The second thread in the episode is about what it actually takes to build something that scales. Erik's 90/10 framework — 90% of budget and effort to scalable, repeatable marketing, 10% to viral — runs counter to how most first-time founders allocate attention. He's seen viral moments generate $10 million in revenue, trigger infrastructure build-out, then vanish — leaving a company with overhead designed for a spike that's already gone. He's equally direct on the executive team side: coming out of COVID, he identified a specific type of stagnation — people who kept referencing the good old days instead of building toward what came next — and made significant changes. The quality he says is hardest to find, and most essential to keep, isn't talent. It's the specific kind of grit that lets someone miss a goal, absorb it, and keep charging. Roland observes that the measurement problem Erik describes — having access to data without knowing what it means — mirrors a pattern he sees consistently in SaaS companies at the $3M–$15M stage. Founders at this stage have typically invested in reporting infrastructure, but the cadence and frame of the reports were built around what's easy to pull, not what reflects their actual sales cycle. The result is a false confidence in the numbers that makes it harder, not easier, to allocate well. Erik's experience — and his data across 5,000 brands — suggests this isn't a scale problem. It's a framing problem that shows up at every stage. Key Moments: 00:02 — Erik's opening claim: AI is mostly regurgitating the internet — and why that's a bigger problem than most companies realize 02:13 — The ROAS fallacy: why comparing this month's spend to this month's revenue will make you cut your best campaigns 04:39 — What actually happens when you scale ad spend from $1K to $5K daily — and why the numbers look broken even when the strategy is working 06:40 — How Hawke uses AI internally: augmentation over automation, and why the "army of 20-year-old interns" framing is more accurate than the hype 12:16 — CEO alignment vs. optimization: why marching in the wrong direction together often beats optimizing in five different ones 15:42 — The 2024 executive team inflection point: what Erik changed, who he kept, and the specific behavior pattern that triggered most of the exits 18:44 — The honest pitch Erik makes to every new hire — and why he deliberately screens out people who find it unappealing 22:28 — The 90/10 marketing rule: why scalable and repeatable should get 90% of your budget, and what happens to companies that chase viral instead 23:39 — The viral sugar rush: how a $10M viral moment can cost a company $10M in infrastructure it no longer needs 25:02 — The Forbes 30 Under 30 moment that made Erik feel, for the first time, like he was competing in the thing he was built for — Hawke Media is offering listeners a free marketing audit. This is most useful for growth-stage companies that are spending on marketing but aren't confident their measurement approach is reflecting their actual sales cycle. https://hawkemedia.com If you're spending on marketing but not sure whether your reporting frame is set up to match your actual sales cycle, this is the kind of problem Midstage works through directly with founders at the $1M–$50M stage. mdstg.ac/drag-erase #MarketingMeasurement #StartupMarketing #OutsourcedCMO #MarketingAgencyScaling #BreakthroughAIOperators

    Most Companies Are Measuring Marketing Wrong — And AI Won't Save Them | Hawke Media | Erik Huberman | EP 202
See All (33)

Trailer

About

Breakthrough AI Operators is a podcast about how the best startup founders are reinventing how their companies work. Not AI hype. Not vendor pitches. Real operators who've rebuilt significant parts of their business around AI — and can talk honestly about what worked and what didn't. Hosted by Roland Siebelink and Doug Miller, co-founders of Midstage Accelerator (7 unicorns built between them, 100+ leadership teams scaled), each episode features a founder who's achieved a genuine step-change breakthrough in how their company operates. These aren't productivity wins or tool adoption stories — they're companies that are structurally different because of AI. If you're a founder at a 20–300 person company actively figuring out what AI means for your operating model and competitive position, this show gives you real stories from people in the field — not consultants theorizing from the sidelines.

Information

  • Creator
    Roland Siebelink
  • Years Active
    2025 - 2026
  • Episodes
    33
  • Rating
    Explicit
  • Copyright
    © 2026 Roland Siebelink
  • Show Website
    Breakthrough AI Operators