A common question that investors ask startup Founders in pitch meetings: Can this company be a Billion-dollar business?

It’s a difficult question to answer in the early stages, though it is easy to parse answers that are directionally correct from those that are strategically flawed. What the question is actually intended to reveal is how the Founder sees and how they describe their path forward.

Exceptional Founders see what others do not and are unafraid to go in that direction.

This is true of startup Founders and investors alike. Elad Gil — most recently co-founder of Color Genomics, previously co-founder of Mixer Labs and responsible for corporate strategy at Twitter and early mobile at Google — is also flat out one of the best angel investors/advisors in the startup community. His portfolio includes Airbnb, Airtable, Anduril, Brex, Checkr, Coinbase, Flexport, Gitlab, Gusto, Instacart, Opendoor, PagerDuty, Pinterest, Samsara, Square, Stripe, and Wish.

He has a kind of sight that makes the nonobvious appear obvious.

This essay unpacks our recent NFX podcast conversation where we examine the counterintuitive —  and often contrarian — patterns that he has observed in many of the world’s best known Billion-dollar companies and their decision frameworks along the way.

The $BQ – Billion Dollar Question For Founders

Something investors love to ask start-up founders: Can this be a billion-dollar business? How do people answer? When you saw a company like Stripe early on, or Airbnb early on, what did you see that let you know that that market was that big?

• It’s a really tough question early on because if something was very obvious that it’s going to be a massive business and market, everybody would already be doing it and there’d be no opportunity for a start-up.
• Definitionally, a start-up has to be doing something a little bit not obvious.
  It’s hard to estimate TAM. Often when people estimate it, they use a BS number where they say: “Commerce is $20 trillion, and if we capture just one sliver…”
• It’s really hard to actually know what the true size of a market is until you’re far enough along that you’re seeing customer adoption and you kind of extrapolate from there.
• You can also underestimate some of these things. So, for example, when I invested in Stripe, which was at the series A and I think they were only eight or ten people, I thought: “Oh my God, it’d be an amazing success if they were worth a few billion dollars.” Now I think it’s going to be a multi-hundred billion-dollar company over time.
• I extrapolated the future growth of the internet, but not enough.
• At the time, like 10 years ago, there were only 5 or 10 $Billion companies, right? That was actually a very, very, very rare milestone. So even imagining something could be $5 billion was a stretch for a lot of things.
• So I think people tend to think about sizing in both directions, right?
  • You tend to underestimate the true game-changing breakout companies at every stage, and that’s why they always look expensive at the time and then cheap in hindsight.
  • Then you tend to overestimate all the other companies in the ecosystem because you apply a view of, “Hey, this is the next Airbnb or this is an extract of things.” For example, half of the unicorns, the billion-dollar market cap companies are not really worth a billion dollars, right? They’re probably not going to end up there, but the other half will, and a subset of that will far exceed what people think.

10X Markets Mean More $B Companies Than Ever

• Most early-stage investors would say the thing that they care about most is the founders. And obviously, founders are incredibly important to a company. They really drive the success of it. I started two companies myself.
• But I think the market is even more important because I’ve seen great founders repeatedly get crushed by a terrible market. And I’ve actually seen some pretty mediocre people do incredibly well if there’s very strong product-market fit.
• Sometimes the company almost runs on its own, irrespective of what the founders do, as long as they have enough of an advantage or there’s a network effect that will sustain them.
• So ultimately my focus is on product market.
• Technology has become such a big force in society and all the biggest companies by market cap are now technology companies. If you look at every single metric, technology markets are at least 10X, if not many tens of times bigger than they were just 10 years ago. If you look at internet usage, time spent online, the number of people with access to the internet, the penetration of e-commerce, et cetera.
• So by every metric standard, markets have literally grown 10X. That means a $10 million niche business is 10 times bigger and is now a $100 million medium-sized business, which means it’s a billion-dollar market cap company.
• So suddenly there are so many more billion-dollar companies that can be created.

“Every Startup Needs To Have A Single Miracle”

• A bad answer to the $Billion dollar question is a strategy of: “We’re starting off doing this one thing, but we’re going to pivot later into this other thing, based on the momentum of the first thing and therefore we’re going to be big.”
• So I’ll give you a bad example from years ago, people used to say, “Well, I want to win in events. So I’m going to start a Yelp or a local competitor so that I get enough liquidity of people who are interested in restaurants and events and then I’m going to pivot into an events product.” But that’s such a convoluted path. Just start with the product you want to make.
• You should just go after the target or prize that you really want to begin with, and figure out what’s the single obstacle and miracle related to that.
• Each startup needs to have a single miracle. A single miracle that it needs to experience or just one giant obstacle that you identify you need to overcome to be successful. If you have more than one, you have compounding small odds and that means you’re very, very likely to fail.
• A startup miracle is the key difficult thing you need to pull off for your startup to work. It could be hitting product-market fit, a key business deal, a specific regulatory change, or the like. If your startup needs zero miracles to work, it probably isn’t a defensible startup. If your startup needs multiple miracles, it probably isn’t going to work — with every miracle, you are multiplying in another low probability event to get an even smaller expected outcome.
• Entrepreneurs often fail when they define an end goal and then come up with a multi-step, multi-miracle-driven logical chain to get there. So if you have multiple miracles that you need to overcome you’re most likely to just fail outright. It’s just too hard.
• James: Smart people, unfortunately, tend to see all the connections between things and they see A then B then C, and then you have failure from a series of cascading miracles that need to happen in order for it to work, which as you pointed out is too many.

Don’t Fear The Single Dominant Vision

Maybe the difference between a company that really meets its potential and one that falls short is this: singular individuals as founders who really take things to the next level.

• Reed Hastings, as an example of a founder who effectively took a company to a level that probably very few people, if any, have made it to. Or Elon Musk with SpaceX.
• Usually, it’s a marriage of “some things are working and I have resources” together with “now let me apply those resources and directions that most people would be scared to go.” Going in a direction that others are scared to pursue is generally the pattern of successful high scale growth.
• Usually, when I see companies blow up in a bad way, it’s because of founder dynamics. People will sometimes start companies with friends and they won’t clarify decision-making — they’ll say we’re equal co-founders, equal in equity, et cetera. But I think ultimately every company needs a final decision maker and you need somebody who can just make a call and then everybody rallies around that call and can move forward.
• I think one of the big myths of Silicon Valley is that every founding relationship needs to be equal. People confuse equal equity with equal input into direction or equal exposure externally, or equal whatever it may be. That gets really over extrapolated.
• If you look at the biggest companies on the planet, many of them had very unequal co-founding relationships. Amazon was really just Jeff Bezos. At Apple, Steve Jobs was dominant. At Facebook, no question about the primary decision-maker there. It used to say Mark Zuckerberg production on every Facebook page. Scott Cook at Intuit was really a sole founder driving it.
• If you look at almost all major breakout companies, it tends to be an unequal relationship.
• Maybe the counter-example is Google. But most, very large market cap companies ended up having somebody who was dominant running it in one form or another.
• Of course, every startup is a team effort, so it’s not like any individual will build everything and do everything.
• But you need somebody who has a clear view of where the company’s heading and that singular view will drive things.
• Many times it’s more important to be able to make a decision, even if it’s wrong and move forward and quickly realize it’s wrong than to spend a long time debating the decision.
• James: At NFX, we always say that each of these startups is a work of art. It’s got to have the fingerprints of somebody, it’s got to feel like something special and unique. It can’t make most of the decisions by committee, because otherwise it just feels beige and it’s not really that much art.

You Know That Stage When Everything Starts Breaking?

There’s tons of advice for early-stage companies in terms of what to do, but there’s very little written for these growth stage or break-out companies.

• Of course, going from zero to one is incredibly hard.
• But I’d almost differentiate between the people who go from zero to one, and the people who go from one to a hundred — there are very few people who can effectively do both.
• The part that tends to have highest complexity and where founders need the most help is when a company really starts breaking out. Suddenly, you’re going from 20 to 100, to 500 people and you’re hiring for roles you’ve never dealt with before. You’re internationalizing, you’re adding multiple product lines, you’re buying other companies, you’re managing board members. You’re doing all this stuff and it’s all happening at once, and everything starts breaking.
• Until you put in place a strong executive team, things will continue to break, and then once you have a bunch of great execs, it tends to stabilize because then you can delegate big chunks of this stuff to them.
• A lot of it boils down to, can you hire a few really great people to anchor the rest of the team as executives? And then can you put in place the minimal amount of process to make sure that things don’t break apart.
• Sometimes a company can get huge just out of sheer and natural momentum. That’s eBay. They never really had an act two of any sort. They bought PayPal and they bought Skype, which were smart acquisitions, but then they spun them off. So, eBay today really isn’t that different from eBay from 25 years ago. So, that’s a good example of a company that never really had a second act and therefore is a massive company because the market is so big.
• Then, there are companies like Google building Waymo, I think Stripe is in that category of company in terms of eventually, they’re just going to be this behemoth.
• The characteristics driving behemoth growth are driven Founders who are continuing to question and innovate things, they’re still reasonably hands-on, they’re empowering their teams to try new things.

Innovation Is All About Timing

• Companies that innovate early to get to a second product line, tend to do that often and build the muscle.
• Companies that innovate late actually never innovate again. If you take five, six, seven years to launch your second area, it usually means you’re not going to ever come up with anything else.
• Caution: Very innovative founders sometimes innovate in places where they really shouldn’t, because it’s both kind of a waste of time, but also, the things that work actually work pretty well. So, it’s this balance.
• On the product side, you definitely want to be constantly re-inventing and thinking of those new forms of distribution, et cetera.
• But you don’t need to reinvent all your HR practices, or you don’t need to reinvent financial accounting. That tends to backfire.
• A lot of technical or product-driven founders, and I’d include myself in that, tend to over reinvent sales and go to market. I actually think sales is a discipline that has existed for decades and it actually works quite well. All that stuff really works if you’re doing top-down sales, or even if you’re doing bottoms-up and augmenting it. You don’t need to reinvent this whole area.
• Don’t waste five years trying to reinvent everything.

Founder-to-CEO Transitions: Hire Another Founder-Type, Not A Stabilizer

• The place where people screw it up on the way to $B companies is a lot of founders, when they leave the company as CEO, they’ll promote the person who is the perfect complement to them to become the CEO.
• So, that’d be like Tim Cook at Apple. You really appreciate that operator skillset.
• In reality, maybe what the founders should be doing is hiring somebody more like them to become the next CEO.
• You kind of need that person who’s hungry and paranoid and scared, and willing to try new things and destroy their own business. That’s not the operator. The operator is the stabilizer, their whole career has been stabilizing.
• So, it’s really interesting to see this pattern where CEO transitions keep going bad because they keep hiring that non-entrepreneurial person or that very operation-centered person who’s the perfect compliment, but again, they need somebody who’s more like them, rather than somebody who’s different from them.
• Stabilizers often come in and they invoke conventional thinking, and often, these first-time founders are successful because they’ve broken with conventional thinking.

Optimize For Every Possible Unfair Advantage

• If you look at every single industry on the planet, there are clusters in the regions where everybody working on a specific thing is in that region. So for finance, it’s New York or Hong Kong or London. For the movie business, it’s Hollywood or Bollywood, et cetera. In every other industry, people find clusters really natural and they tell you, “Hey if you want to make movies, you should really go to Hollywood.” They never say, “Hey, if you want to make movies, you should really go to Austin.”
• If you think about it, you can write a movie script from anywhere, you can edit it digitally from anywhere, you can score the film from anywhere. You often shoot offsite anyhow, in terms of the location. So why does Hollywood exist as a cluster? It’s because there are lots of network effects to a bunch of people working on the same thing, getting together in terms of supplier relationships, partnerships, capital, hiring, everything else. That’s also true for tech.
• So I don’t think the Bay is going anywhere. I think when things snap back after COVID, the Bay Area at least will continue to be the primary US-based technology hub. It’s always possible it’ll lose some sort of market share equivalent in terms of technology companies, but I don’t think it loses half, maybe it loses 10%.
• In general, my advice would be to go to a technology cluster. If I were to go to one cluster post-COVID, it would be the Bay Area. And then after that, it’d probably be New York and then LA just in terms of size and density, and then for people doing interesting things.
• But again, there are these smaller clusters sort of forming elsewhere. You can always start a company elsewhere, right? I mean, Shopify is a giant, and there are other giant companies based outside of the clusters. It’s just a lot harder in my opinion.
• It’s a question of what sort of advantages do you want to have. In a cluster, you’re in-the-know earlier on new distribution hacks, or you have a bunch of customers you can sell to locally if you’re building a SAAS product, right? A lot of the SAAS companies that succeed have other startups adopt their software first. And so if you can get your first 10 customers much more easily, that actually makes a huge difference in terms of the rate at which you iterate and learn and start to have revenue and all the rest of it.
• I do periodically analyze where is all the unicorn market cap, which is sort of backward-looking, and over the last 15 months about 180 new unicorns were generated, which is insane. About a quarter of them globally are in the Bay Area.

You can listen to the podcast conversation here.