The Best Way To Kill Your Startup & What I Learned Doing Just That

The Best Way To Kill Your Startup & What I Learned Doing Just That
Photo by Brendan Church / Unsplash

A year ago, I wrote this internal memo to our team at XMTP Labs. It's something I think about a lot, and even as we've grown and come a long way since a year ago, I still think about this almost every day. Here's how 13 years ago, we blew $35m dollars at Zaarly, failed miserably, and how we are doing it differently this time around.


Oct 20, 2022:

Hey team,

I hope everyone had a good weekend.

I want to share a story and a fear I have for where we are at as a startup. I want to start with a quote:

“More Startups die from indigestion than starvation” - Bill Gurley

In 2010, I was leading product at a company called Zaarly. “Local, mobile, social” was the buzzword of the time. I guess rightly so, since Uber, Postmates, DoorDash, Airbnb, and many other massive local marketplaces came out of this era. An underlying shift in technology opens up a once-in-a-moment opportunity to build something that can impact the entire globe.

Zaarly was right there. We were a local marketplace to ask for anything you want. We had raised $35m+ in venture funding. Hired 50+ people. Had incredible offices in San Francisco. A bunch of cool people invested. Smart people joined our board. We had everything we needed, and more.

But we didn’t. We died.

It was the hardest years of my entire life. Both personally and professionally. We were at the right place at the right time, but we completely missed what I thought would be the biggest opportunity of my life. None of the cool or successful people mattered. It was sure good for validating our ego, not building something meaningful. I personally lacked the self-awareness to change what wasn’t working, quickly.

We didn’t obsess over the customer experience or nail an insanely specific use case that our users loved every single time.

I even remember our ego in the office. we’d say things like “Ha, Wag only does dog walking? Postmates only does food delivery? Our technology is so much better. That’s just one vertical on our platform, we are thinking so much bigger and broader. Our platform does so much more.”

"Our platform does so much more" might be the language to realize you're going to die as a startup.

Turns out, dog walking was a billion-dollar opportunity. Postmates was too. Babysitters. Almost everything. Pick a vertical, I'll show you a $B company that was created. We couldn’t have picked a small enough vertical. Like most things in disruptive technology, the market size is exponentially bigger than we can comprehend today.

It’s easy to convince ourselves that going broader means going bigger. I’ve almost never found this to be true in software. Even a protocol.

We died because our strategy was bigger than our vision.

Our vision was massive, to create a new local economy or something ridiculously huge like that. I still remember the hubris in our pitch “Have you ever said you want money? You’re welcome.” Or something completely fucking ludicrous like that. Blind arrogance.

I’ve actually found that great strategy is about stripping away everything that’s not critical to what people need and getting insane value from today, and then and only then can realize our massive and ambitious vision.

Our ability to get specific with partners and deliver something that they truly love and being critical to solving their problem is the thing. A good friend of mine told me once;

Build smaller and better to go bigger. It's counterintuitive but it's critical. Until the specific thing you're building is loved by people, don't do anything else. If it isn't really working, kill it before you expand it. - Robert Stephens, Founder Geeksquad

If we look at SMS, the protocol may be broad but the use case that made it explode was as simple as 2FA login flows for most of the mobile app world. We may be building a general protocol but we are searching for the killer use case with our partners and making sure our protocol delivers on the promise of whatever this use case will be.

Interoperability and a simple use case made SMS. Fun fact: 20% of Twilio’s revenue at IPO was because of the login flow of WhatsApp. That seems insane but highlights the point.

Being open and honest about our current reality while being insanely optimistic about our opportunity is what I’ve learned to value most and the teams who win.

To be crystal clear - there are two big gaps that keep me up at night and are causing me to draw parallels to Zaarly:

  1. We do not have product market fit
  2. We are not junkies of our own product to ensure it works great for us and our partners. We are lost between our own app and serving developers. We must get clear on who our customer is, and fast. (Update: We killed our app, and focused solely on developers, and it was the best and most critical decision we made)

My biggest fear is having the opportunity of a lifetime and failing at it again. That’s why I’m sharing this with you because we aren’t going to do that, we are going to win this time.

  1. We have a lot of capital for pre-product market fit, let’s not let this bloat us. We should focus on small teams, fast iteration cycles, and being radically self-aware about what’s working and what’s not. More people doesn’t make companies go faster. Small groups of people focused on specific things do. Scaling globally comes from finding the thing not scaling everything. Millions of people can understand one thing, not everything. Let’s go find it.
  2. Let’s make sure our partners love using XMTP, and let’s be sure we LOVE using XMTP, too. I don’t think we’ve gotten here yet but I believe we can. There are too many bugs, difficulties to integrate, and unknown unknowns that cause a lot of friction with our partners. How can we make it simpler faster and better? How can it be a magic moment the second they integrate? Let’s find that and make that happen. Let’s make magic. Most importantly: talk to partners, more. It’s never enough.

More startups die from indigestion than starvation. Said more simply: Money does not equal speed & money does not equal Success.

Speed = Success. It’s the single most important thing to remember because it’s the only thing that a startup has that an incumbent doesn’t. It’s the learning loop or iterations.

We find product-market fit through speed and iteration with partners, not internal debates or adding more complexity. The learning loop is how we know what to do next. We win by making it radically simple for our partners by doing the hard work to make complexity simple.

This moment is one of the biggest opportunities I’ve ever seen in my career. I’m so glad I feel like I got another shot.

I’ve never felt luckier to have landed in this moment with the right team, right insights, and right investors to be able to capitalize on it.

Let’s come back to today.

It’s happening and we are perfectly positioned to capitalize on it.

I’ll reiterate: What matters now is the speed of iterations, speed comes from smaller teams, and the self awareness and focus to make sure our partners love using XMTP.

Having 10 partners who LOVE and go all-in on XMTP is much more valuable than having 50 partners who think it’s cool and “kinda” use it.

People sometimes ask me “How do you know if people actually love our protocol?”

I’ve found that if we have to ask, the answer is that they probably don’t, yet. Same with product market fit. If you're wondering if you have it, you probably don't. You know when you have it.

I hope my lessons from the past can help us stay humble, stay nimble, and realize that speed is the single most important thing in a startup.

I’ve never believed more in this team, this market, and this opportunity. Let’s go make it happen together.

Shane Mac - CEO & Co-founder XMTP Labs


We now have over 500+ developers building with XMTP, Coinbase Wallet launched, and over 4.5m identities have joined the XMTP network. We are just getting started but there's such a big opportunity to build a private and better communication network for the world. We are hiring across all teams, come join us. https://blog.xmtp.com/careers/

Subscribe to Shane Mac

Sign up now to get access to the library of members-only issues.
Jamie Larson
Subscribe