Olivier Truong

The Origins of Famous Startups

Where do ideas for startups like GOAT and Instacart come from?

I like studying the origins of successful companies. Going through the existential founder journey, I get a mix of inspiration, motivation and comfort out of these stories. Inspiration because they help me understand where ideas come from. Motivation because bringing an idea to life is a dream I live for. Comfort because some stories tell of struggle and sorrow and knowing that I'm not alone helps me keep going.

I'm writing a two-part series on the anatomy of ideas:

  • How famous startups got their idea (this post)
  • What to do today so that one day you'll have that special idea


Over the last year I've listened to many a podcast and read many books where founders recount the early days of their companies. Here are the 7 ways that

Here are the 6 ways that 50 well-known companies found their idea:

  1. Start with a framework, brainstorm lots of ideas, evaluate them rigorously and only pursue if it feels right. (Netflix, Glassdoor, Diapers.com)
  2. Side project / side hustle turned business: the Airbnbs rented their living room and air mattress to make rent, the founder of Ring wanted to hear the doorbell from his garage. Faire, Allbirds, DoNotPay
  3. Found problems worth solving by working on a previous idea that failed (Retool,Slack, Descript, Discord, Sendbird, Segment, Notion)
  4. Idea emerged from hobby (GOAT, Calm, Strava)
  5. Noticed business opportunity by running into problems at their previous job (ZipRecruiter, Zapier, Calendly)
  6. Looked for ideas by talking to business owners (Doordash, Alto)
  7. Pick an emerging technology and try many ideas in that domain until it works (Figma)

Editor's note:

Some people will tell you that finding a startup idea is easy. Sans obvious but difficult to execute ideas like self-driving cars, I don't think that's true. Finding a good startup idea is hard - hard in the sense that it is unpredictable. It's easy to be fooled into believing anyone can succeed because of stories like Snapchat or Patreon - disappearing photos.. pretty html on top of Stripe?? You'll never hear from people who tried hard but never made it because the power of the internet only surfaces winners. I call this out to not mislead readers into believing there is a formula for this stuff. A lot of luck is required. In the next post, I'll explore what one can do to increase the amount of luck they attract.

Substack - Hamish McKenzie and Chris Best had just left Kik, the messaging app company. Hamish was/is an accomplished writer working as a lead in communications at Kik and Chris was one of the co-founders of Kik and CEO. While writing an essay about online advertising, the news, tech, sensationalism and divisiveness, Chris sought out Hamish for feedback and ideas for how to solve these problems.

[At this point, it's unclear how intentional they were about starting a company and how long they brainstormed until coming up with the idea for Substack.]

Out of these conversations came the idea for Substack, inspired by Ben Thompson's paid newsletter, Stratechery, a popular newsletter in tech circles. Substack's offering would be simple: allow writers to start a newsletter and make money from subscriptions.

[They convinced Bill Bishop, the author of a popular free newsletter about China, to be a very very early user and to help shape what Substack would become. It's unclear but I assume that Hamish, a writer himself, had some history with Bill. No doubt, these connections were critical to make the product as good as it is now but also to seed it with literary cachet to get early momentum going for the brand.]

Netflix - It's the year 1997, the internet is all the buzz, and two former tech executives brainstorm ideas for a new company for months. Hundreds of ideas. Personalized shampoo by mail, personalized surfboards, pet food. They know it has to be something consumers buy frequently. Suddenly, inspired by a sleepless night watching Aladdin with their three-year old, one of the men suggests video tapes. They aren't immediately convinced but they look into it.

There were a number of reasons to believe they couldn't compete with Blockbuster selling VHS tapes.. until they learned about a new emerging technology called DVDs.

Discord - Discord's first taste of success was on May 13, 2015 when someone posted about Discord on the Final Fantasy XIV subreddit. In 2014, Jason Citron and Stan Vishnevskiy were building an online mobile multiplayer game but it wasn't working. They liked the chat and voice system they'd built. Realizing that gamers were still using TeamSpeak, they plotted to offer them something better.

[At this point, it seems that the first version of Discord is being built out in parallel with continuing development on their mobile game. Of course, they already had a full team of developers.]

So they built a web-based voice and chat application. They only had 10 users at some point - it wasn't sticking. It took off when they improved call quality and added roles and moderation features. Their first surge of users was that day on May 13, 2015.

GOAT - Daishin Sugano and Eddy Lu had just shut down their startup Grubwithus - they had $2.5M left. Another year goes by where they struggle to pivot Grubwithus into something that works. $1.5M remains when Daishin gets scammed buying a retro pair of Air Jordan 5 Grapes on eBay, a purchase motivated by the nostalgia of his dad buying him his first pair of Jordan shoes when he was 10 years old - that exact model. Disappointed, maybe even angry, Daishin tells Eddy about it and they start talking about the experience. It turns out Daishin had been a long time sneaker enthusiast and Eddy pulled on that lace. Further questioning made it apparent there was an opportunity.

Figma - In 2012, Dylan Field received $100k from the Thiel Fellowship to drop out of school and start a company. His cofounder Evan Wallace, an endowed programmer, had graduated that year and decided it would be more interesting to work with Dylan than to join a company. Dylan's application to the Thiel Fellowship had described a plan to build drone technology but this was quickly dropped when they decided to focus on WebGL and its opportunity to change the web. Dylan had interned at Flipboard and experienced the power of design tools, while Evan, a programmer's programmer, tinkered with WebGL. So they tried a lot of things. A few failures in, they find themselves building a meme generator and a mixture of despair and embarrassment sets in when they reflect on dropping out of Brown to do this. Next they consider building a photo editing tool - again, in the browser. They downvote the idea when they see the trend of photography going to mobile and thus photo editing going to mobile, not the desktop browser. But nine months in, they start working on a photoshop for the web and that's when the vague outline for Figma starts to form. It wasn't until later that they decided to niche down on interface design - riding the wave of Product Design becoming ubiquitous as every business goes digital.

DoorDash - Tony Xu and Evan Moore were second-year B-school students brainstorming ideas for a company in 2013. Evan met Stanley Tang, a third-year undergrad, through a CS class and asked if he'd like to work on something - they didn't know what yet. Stanley brought in his friend, Andy Fang, another CS undergrad. Now four, they decided they had a common interest in helping small business owners. They went door-to-door speaking with local businesses to learn about their operations. One of their failed ideas was a marketing attribution survey via iPads placed at retail point-of-sale. One day, they visited Chantal Guillon, a macaroon shop in Palo Alto, to ask about the business and to get feedback on a prototype they'd built for whatever idea they were working on at the time. During this meeting, Chloe, the owner, pulled up this thick booklet. Inside were pages and pages of delivery orders, many of which, she explained, she had to turn down - she didn't have delivery drivers. She was delivering orders herself.

This was the lightbulb moment. They interviewed other shops and restaurants and learned that most didn't deliver and those that did deliver - they hated doing it!

[Stanley Tang says they spoke to an additional 200 businesses to validate the delivery problem. If that's true that's an impressive level of rigor that most startups don't apply especially pre-product.]

Armed with this well-defined problem, they quickly launched a basic static one page website at paloaltodelivery.com with a phone number and several PDF menus, alongside a small adwords campaign. Hours later they got their first phone call.

Strava - An idea 10 years deferred. In 2008, Mark Gainey and Michael Horvath looked to start their second company. They sought out spaces they could be passionate about. They researched sustainability and getting rid of plastic water bottles, but that's as far as it went - research. They went back to an idea they'd kicked around in 1996, over a decade prior. They wished to recreate the competitive environment of being on a sports team - being pushed to get better. You see, both men had been college athletes. Seeing the rise of Garmin fitness trackers equipped with a GPS function, they saw their opportunity. They hired a team of six and created a prototype. They assembled a group of 20 cyclist friends, gave them Garmin trackers with instructions to upload their data daily via their web app. They split them up into two teams: west coast vs east coast. Over email they coordinated regular competitions such as post the fastest 5k in the next 24 hours. They loved it.

Diapers.com - Marc Lore started a sports collectible company in 1999 and sold it for $5.7M in 2001. Four years later, looking for his next idea, he searched keywords in Google. At the time the search engine showed you how many times a keyword had been searched. The lightbulb moment hit when he searched "diapers" and saw that it had been searched 200,000 times in the last month. He realized that you couldn't buy diapers online, not even Amazon. At least not at reasonable prices. He was 34 at the time and already a father. He thought, "Everyone needs diapers and it's a pain in the butt. If I can deliver them overnight, great service at good prices, isn't every mom in America gonna want to do that?" There were objections. He quickly learned diapers are a loss leader at brick and mortars so the economics would only get worse if they had to be shipped. His brilliant decision: use the same loss leader strategy - but online - with the unlimited shelf space of the internet. It worked.

Alibaba - Jack Ma's story is legendary. Here's an abbreviated telling of his path to founding Alibaba. Jack learned English at a young age. In 1994, he starts a translation agency in his hometown of Hangzhou. Some time later, the nearby county government asks him to assist in a dispute with an American company over some construction project. He ends up in Seattle and a friend of a friend shows him the internet. He finds a lot of content but nothing related to China. Intrigued, he asks his new American friend to help him make a basic page for his translation agency. To his surprise, he starts getting emails from all over the world. That's when he gets the idea for a new business: an online, English directory of Chinese businesses looking for overseas export partners.

[If it's not obvious by now, Jack Ma was intent on building on a successful company before he even knew the internet existed. This was a mind prepared to see opportunity.]

He called his new business China Pages. Unfortunately, it didn't do well. The website couldn't facilitate payments online so it was limited in the amount of service fee it could charge.

The rest of the story is just as fascinating as the first part. In short, Jack Ma leaves his company China Pages, he works for the government building its own version of China Pages, he eventually leaves and tries another private enterprise with Alibaba and, as you know, it works.

Instacart - Apoorva Mehta was going through one of the lowest points in his life, finding himself crying in bed, when he came up with the idea for Instacart. Two years prior he'd quit his job as a software engineer at Amazon. He'd moved to San Francisco and was sleeping on his friend's couch. It took him 2 years and 20 failed ideas until the idea for Instacart came to him. Trying to come up with another idea, he remembers reflecting on how everything he did was online. He shopped online, he met people online, he watched TV online. The one thing he hated doing, grocery shopping, he still had to do the same way people did when grocery stores were invented.

[Apparently it wasn't just that grocery shopping was inconvenient in SF without a car. Back in Canada, Apoorva had despised waiting for the bus in the cold with heavy bags pulling on his frozen fingers.]

Apoorva fell in love with the idea. He describes it as a crazy high in the midst of a very sad time in his life.

How do you prepare to start a startup in the future if you don't have ideas now? This question requires another essay to unpack. Here's a teaser anyway. I say, first, fail many times - do this as early in life as possible. Experience the sharp pain of failure and internalize the consequence of fuzzy thinking. Then, realize that ideas emerge when we do interesting things and, critically as important, when our mind is trained to look for opportunities. You, too, can be Daishin Sugano if you're prepared to see the opportunity start a sneaker marketplace when you get scammed on eBay.