There's No Such Thing As A "Best Pitch"
Entrepreneur Office Hours - Issue #339
Inside the Office
I recently worked with a cohort of young entrepreneurs preparing to pitch their ideas. As always, I loved the coaching part.
I am in my happy place when I’m helping founders sharpen their story, clarify their thinking, and find the thread that makes an audience lean in. What I don’t love is judging.
In fact, being a judge is one of my least favorite jobs to be done.
I was reminded how little I enjoy judging when I was recently asked to judge a pitch competition sponsored by Inc. Magazine for high school entrepreneurs. The founders were eager. Their ideas were creative. But what jumped out at me most wasn’t the pitches themselves.
It was the other judges.
As we discussed the finalists, it became clear that we weren’t all evaluating the same thing.
Some judges rewarded great storytelling. Others cared more about opportunity size. Some cared most about execution and traction. Others were drawn to founders who demonstrated deep expertise and clearly knew their subject matter inside and out. And others cared more if the imaginary revenue projections were math-ing…
The same pitch could land completely differently depending on who happened to be listening and what their ear was listening for.
Which leads me to something every founder (and professor) eventually learns:
There is no universal definition of “best.”
We like to believe that competitions, investors, customers, and markets are objective. That if we just work hard enough, polish enough, and put in the reps, there will be a clear winner.
But entrepreneurship doesn’t work that way.
The “best” pitch isn’t a formula (although there are some very important boxes to check). It’s not a spell you’re casting on others. It’s a conversation between a founder and an audience, and every audience brings its own biases, experiences, agendas, and hard-earned scars.
An investor who spent years losing money on marketplace businesses hears a marketplace pitch differently than one who made a fortune investing in them.
A customer who has lived with a super specific problem for years sees huge potential where others see a narrow niche.
A judge who values potential will listen for different signals than a judge who values traction.
The goal, then, is not to create a pitch that every person will love.
That’s impossible.
The goal is to create a pitch so compelling that the right people can’t ignore you.
The founders who ultimately stood out in the competition didn’t necessarily have the most beautiful slides. They weren’t always the most charismatic, either.
They had ideas grounded in insight and evidence. They were clear and simple. They understood the problem deeply. They had practiced. They could articulate why the world should change and why they were the people to make that change happen.
Most importantly, their ideas felt inevitable.
They answered the questions the audience is asking, whether they say them out loud or not:
Why this?
Why now?
Why you?
As founders (and humans) we spend far too much time worrying about whether we’re the best at something.
There is no objective “best” pitch, founder, or startup.
There are only people trying to convince other people that a different future is possible, the time for that change is now, and you are the one person on earth to bring this vision to life.
-Amy
Worth Your Time
Sam Altman may have just given us a preview of the next era of entrepreneurship. In this article from Inc. Magazine, OpenAI is offering $2 million in OpenAI API tokens to startups in Y Combinator’s Spring 2026 batch in exchange for an undetermined amount of future equity. Not cash for equity. Not office space for equity. Not accelerator programming for equity. Compute for equity. That may sound like a clever startup financing idea. I think it is much bigger than that.
For decades, the answer to the startup resource question was obvious. Founders needed capital. They needed money to hire engineers, build software, buy infrastructure, acquire customers, and survive long enough to find product-market fit. But AI changes the math.
If a founder can use AI to write code, generate content, build prototypes, analyze markets, test ads, automate customer support, build internal workflows, and personalize products, then the scarce resource may no longer be only money. The scarce resource may be intelligence at scale. And intelligence at scale requires compute. So is compute becoming the new cash? I think the answer is yes, at least partially.
Cash still matters. Payroll matters. Sales matter. Customer acquisition matters. Legal fees, travel, distribution, data, and real-world execution still matter. But for AI-native startups, compute may become what manufacturing capacity was to industrial companies, what shelf space was to consumer brands, what server capacity was to internet companies, and what oil was to the twentieth-century economy. It is not just an expense. It is an input into possibility.
Would I take the deal? Maybe. But not blindly. If I were a YC founder, I would want to know what I was really giving up. Two percent of a company sounds small when the company is just an idea, a pitch deck, and a few people sitting around a table. It sounds very different if that company becomes worth a billion dollars. I would also ask whether the compute accelerates learning or simply subsidizes waste. Free tokens can make founders sloppy. The goal is not to use more AI. The goal is to learn faster, build faster, sell faster, and get closer to the truth faster.
I would ask whether I was becoming too dependent on one platform. If my product, workflow, cost structure, and roadmap are all tied to OpenAI, am I building a company or am I building a feature in someone else’s ecosystem?And I would ask what OpenAI sees that I may not. If OpenAI wants equity in hundreds of startups in exchange for compute, they are not just being generous. They are buying insight, option value, ecosystem influence, and a front-row seat to the next generation of AI-native companies.
This should also get the attention of venture capital firms. Not because OpenAI is replacing venture capital tomorrow. But because the value stack around startups is changing. Traditional venture capital offers cash, credibility, pattern recognition, networks, hiring help, and follow-on financing. That still matters. But what happens when a platform company can offer something more immediately useful? What happens when OpenAI, Anthropic, Google, Amazon, Microsoft, Nvidia, or Meta can say to a founder: we will give you the fuel to build? That is a different kind of capital.
In the AI era, money is necessary but less differentiated. What is scarce is unfair advantage. The real question for investors may become: what do you bring that helps this company learn, build, sell, or scale faster than anyone else? If the answer is only money, that may not be enough. Will YC companies take the deal? Many will. Of course they will.
Early-stage founders are resource constrained. Equity feels abstract. Compute feels real. Two million dollars of AI capacity could let a young company build more, test more, serve more customers, and try things that otherwise might be impossible.But the best founders will not treat this as free money. They will treat it as high-octane fuel. And high-octane fuel is only valuable if you know where you are driving. The dangerous founder response is, “Great, now we can use AI everywhere.” The disciplined founder response is, “Where can this compute create proprietary learning that compounds?”
The pricing question is also fascinating. Cash is easy to price. A dollar is a dollar. Compute is different. Its value depends on who is using it, for what purpose, at what stage, with what model, and with what outcome. For one startup, $2 million of tokens may be an expensive science experiment. For another, it may be the difference between a demo and a real product. For another, it may unlock an entirely new business model that could not have existed five years ago. So the real value of compute is not just its sticker price. It is its option value.
Compute allows founders to run more experiments, serve more users, train or fine-tune more systems, personalize more deeply, automate more workflows, and compress time. And in entrepreneurship, compressed time is often the most valuable thing there is. If you can learn in two weeks what used to take six months, you have not just saved money. You have changed the slope of the company.The bigger implication is that compute may only be the beginning. If compute can be traded for equity, what else can?
Data can be traded for equity. A hospital system, retailer, university, logistics company, or financial institution with proprietary data may become a strategic investor in startups that can turn that data into products, insights, and outcomes. Distribution can be traded for equity. The hardest thing in many startups is not building the product. It is reaching the customer. A company with trusted customer access may have something more valuable than a check. Workflow access can be traded for equity. Imagine a startup embedded inside a law firm, hospital, manufacturer, university, or sales organization, building AI tools around real pain points in real time. The company providing the workflow becomes an investor through access. Expertise can be traded for equity. Not advisory shares in the old ceremonial sense. Real expertise. A world-class doctor, engineer, operator, teacher, CFO, or sales leader can help train, validate, and shape a product in ways money cannot.
Brand trust can be traded for equity. In a world full of AI-generated noise, trust becomes more valuable. A respected institution or community can reduce customer hesitation and create legitimacy. Talent networks can be traded for equity. Access to elite builders, researchers, domain experts, and operators may become a capital source. Regulatory navigation can be traded for equity. In health care, finance, education, energy, and defense, knowing how to move through complexity is not a side service. It is a strategic asset. Customer commitments can be traded for equity. Early customers may increasingly say, “We will help you build this, test it, and buy it, but we want ownership.” Community can be traded for equity. A deeply engaged audience may be more valuable than an ad budget. Attention is everywhere. Trust is scarce.
That is what makes this story so important. The old founder question was, “How much money do I need to raise?” The new founder question may be, “What resource changes the trajectory of my company?” For some founders, that resource is cash. For others, it is compute. For others, it is data, distribution, trust, talent, credibility, customer access, regulatory insight, or speed. Entrepreneurship has always been about the pursuit of opportunity without regard to the resources currently controlled. But AI changes the resource map.
The founder of the future may not ask only, “Who will fund me?” The founder of the future may ask, “Who can give me the unfair advantage I cannot buy later?” I would not dismiss this OpenAI offer as a publicity move. I would also not treat it as charity. This is a platform company making a strategic bet that the next generation of startups will be built on AI infrastructure, and that owning a small piece of many of them may be very valuable. For founders, this is exciting and dangerous. For VCs, it is a warning shot. For universities, it is a wake-up call. For students, it is an invitation.
The tools are no longer theoretical. The cost of building has collapsed. The speed of experimentation has exploded. The definition of capital is changing. The distance between idea and product is shrinking every day. So the question is not just whether you would trade equity for compute.
The better question is this: If someone gave you $2 million of intelligence tomorrow, would you know what to do with it?
 | A guest post by
|
 | A guest post by
|