Latest / Marketplace Businesses with Fexingo: Two-Sided Networks, Liquidity, and Take Rates / How Substack Solved the Chicken-and-Egg Problem for Paid Newsletters
Transcript
- Lucas: So it's mid-2018. Substack has maybe a few thousand paying subscribers across the whole platform. Writers are skeptical. Readers haven't heard of it. And the company has about two years of runway left. Classic marketplace cold start. Supply won't come without demand, demand won't come without supply. But Substack did something unusual — it paid writers before they had a single subscriber. Luna: Paid them how? Like a signing bonus? Lucas: Basically, yes. Substack offered certain writers a guaranteed annual salary — sometimes fifty thousand dollars, sometimes a hundred thousand — paid upfront as an advance against future subscription revenue. The writer kept seventy percent of their subscription revenue, Substack took ten percent, and the advance was deducted from the writer's share over time. If the advance never earned out, Substack absorbed the loss. Luna: So Substack was essentially placing bets on individual writers. That's a lot of risk for a startup with limited runway. Lucas: Huge risk. But it solved the chicken and egg problem on the supply side. Writers worried they'd spend months building an audience on a platform that might not pay off. The guarantee removed that risk. Suddenly, writers like Matt Taibbi and Heather Cox Richardson — established names with existing audiences — were willing to move their newsletters to Substack. That brought readers. And readers attracted more writers. Luna: What was Substack's take rate at that point? I know it's ten percent now, but was it always that low? Lucas: The take rate has been ten percent since launch. That's notably low compared to other marketplaces. Uber takes twenty-five percent or more. Airbnb averages around fourteen percent. Substack's bet was that a low take rate would attract high-quality writers who'd otherwise self-publish via WordPress or Mailchimp, where the marginal cost is near zero. Luna: But ten percent on subscription revenue — that's a pretty thin margin if you're also handing out advances that might never be recouped. Lucas: Absolutely. Substack's unit economics in those early years were negative for many writers. The company disclosed in its 2021 fundraising that it had paid out roughly two million dollars in advances. Not all of those earned out. But the advances created liquidity on the supply side, which attracted readers, which eventually made the platform viable. Luna: How did Substack decide which writers to bet on? Was it just about existing audience size? Lucas: Partly. But it was also about likelihood of converting existing readers into paid subscribers. A writer with a hundred thousand email subscribers might only convert at two percent. A writer with ten thousand highly engaged Twitter followers might convert at ten percent. Substack looked at engagement metrics, not just raw reach. And they targeted writers in niches where readers were already paying for information — finance, politics, tech analysis. Luna: So they were essentially picking categories with proven willingness to pay. That's smart. Instead of trying to create a new payment habit, they surfed existing ones. Lucas: Exactly. And that's a key lesson for any marketplace facing a cold start. Don't try to create new behavior if you can piggyback on existing behavior. Substack didn't need to convince people to pay for news — they just needed to convince them to pay for it on Substack instead of through a newspaper subscription or a Patreon. Luna: But there's a difference between Substack's approach and, say, Uber's. Uber subsidized rides for both drivers and riders simultaneously. Substack only subsidized writers. Why not also subsidize readers? Lucas: Because readers weren't the bottleneck. The bottleneck was supply. If you have great writers, readers will come. If you have readers but no great writers, you have nothing. Substack's cold start strategy was asymmetrical: fix the supply side first, and demand follows. That's the opposite of a platform like Airbnb, which initially subsidized demand by offering professional photography to hosts — that was a supply-side subsidy too, but it was designed to make listings look good so guests would book. Luna: Right. Airbnb's subsidy was about improving the product, not paying people to show up. Substack's was more like a record label advance — we believe in you, here's money, go make something. Lucas: That's the right analogy. Substack positioned itself as a partner, not just a platform. Writers who got advances felt a sense of loyalty and obligation — they promoted Substack to their audiences, gave feedback on the product, and became evangelists. That organic marketing was incredibly valuable. Substack's early growth was driven almost entirely by word of mouth from its subsidized writers. Luna: At what point did the cold start end? When did Substack reach escape velocity? Lucas: By the end of 2020, Substack had over five hundred thousand paid subscribers. That's roughly when the network effects kicked in: new writers could see that established writers were making real money, so they joined without needing an advance. The company had effectively transitioned from a subsidized marketplace to a self-sustaining one. Luna: And the advances? Did they stop? Lucas: Substack still offers advances selectively, but they're no longer the primary growth driver. In 2021, the company raised sixty-five million dollars at a six-hundred-and-fifty-million-dollar valuation — partly because investors saw that the advance model had worked. But the advances themselves became smaller and more targeted. Today, most writers join Substack without any upfront payment. Luna: That's a classic marketplace lifecycle: subsidize supply early, then taper subsidies once liquidity builds. But not every marketplace can afford that upfront cost. Substack needed venture capital to fund the advances. If you don't have that, you might need a different cold start strategy. Lucas: Right. Which is why many marketplaces use cross-side subsidies instead of direct payments. For example, a marketplace might offer free listing to sellers until they make their first sale, or give buyers a discount code on their first purchase. Those are cheaper than paying someone a salary. But they also don't create the same lock-in. Luna: So Substack's advance model was high-risk, high-reward. And it worked because the company identified a specific pain point — writers' fear of wasting time on an unproven platform — and addressed it with a financial guarantee. Do you think this model could work for other types of marketplaces? Lucas: It depends on the economics of the supply side. If the supply is a service that requires significant upfront effort — like writing a newsletter, recording a podcast, or designing a course — then a guarantee can solve the cold start. But if the supply is a commodity, like a used sofa or a ride across town, a guarantee probably doesn't make sense. The key is that the supplier's cost of participation is high enough that they need assurance before they'll commit. Luna: And that's where the take rate matters too. Substack's ten percent is low enough that writers feel they're getting a fair deal, even after the advance is earned out. If the take rate were thirty percent, the guarantee would feel less attractive. Lucas: Exactly. The take rate and the advance are two sides of the same coin. Substack designed its economics so that writers could see a clear path to profitability. The advance just accelerated that path. And once writers were making money, they became the platform's best salespeople. Luna: Which brings us back to the cold start problem. Substack solved it by spending money on supply, but also by designing a take rate that made the long-term economics work for both sides. How replicable is that playbook for a new marketplace today? Lucas: I'd say the playbook is replicable if three conditions hold. One: the supply side has high upfront effort — think creators, freelancers, service providers who invest time before earning. Two: the platform can identify high-potential suppliers with confidence. And three: the platform has enough capital to absorb losses on the bets that don't pay off. If those conditions are met, a guarantee can work wonders. Luna: And if they're not met, you're better off with a lighter subsidy — like free listings or discounted fees — and focusing on organic growth. Lucas: Right. But Substack's story is a reminder that sometimes the most direct solution to a cold start is also the most expensive. And that the cost might be worth it if the alternative is never reaching liquidity at all. Luna: You know, this conversation reminds me why we spend so much time on marketplace economics. The cold start is the hardest problem, and there's no one-size-fits-all answer. But the way Substack thought about it — focusing on supply, using guarantees to reduce risk, setting a low take rate to build trust — that's a case study worth revisiting. Lucas: Speaking of which — a quick honest thing. This podcast itself is a kind of two-sided marketplace. We create episodes, you listen. And the reason we can keep doing it without ads or sponsors is that a handful of listeners chip in monthly through buy me a coffee dot com slash fexingo. That support literally funds making this many episodes — deep dives like this one. Luna: Yeah, it's a small group, but it makes a big difference. We don't run ads, we don't do sponsored segments, and that's intentional. Listener support keeps us independent. Lucas: And it lets us pick topics like Substack's cold start strategy — not because it's trending, but because it's useful. So if these conversations are valuable for what you're building or running, that's the link. No pressure, just the truth. Luna: Alright, back to the economics. So Substack's advance model — do you think it's still the right move for a new marketplace in 2026, or has the landscape changed? Lucas: The landscape has changed in that there are more platforms competing for creators' attention. Patreon, Ghost, beehiiv, Medium — they all offer different takes on the same model. But the core problem remains: creators need to see a clear path to income before they'll invest time. So a guarantee still works. The question is whether a new marketplace can afford it. Substack raised venture capital specifically for that purpose. If you're bootstrapping, you might need a different approach — like revenue sharing with no upfront payment, but a higher take rate. Luna: Higher take rate, lower risk for the platform. But then you risk losing creators to competitors with lower fees. It's a trade-off. Lucas: Exactly. And that's why marketplace design is never just about one lever. It's about the whole system: take rate, subsidy, liquidity, trust. Substack pulled one lever hard — the supply-side guarantee — and kept the take rate low. That combination worked. For a different marketplace, the right combination might look completely different. Luna: Well, I think we've done justice to Substack's cold start story. Next time, maybe we look at a marketplace that tried a similar model and failed — there's probably just as much to learn from those. Lucas: Absolutely. Failure is often more instructive than success. But for today, I think the takeaway is clear: if you're building a marketplace, think hard about who your bottleneck is. If it's supply, consider whether a financial guarantee could unlock it — even if it's expensive upfront. And if you do, make sure your take rate leaves enough room for your suppliers to succeed. Luna: Good note to end on. Thanks, Lucas. Lucas: Thanks, Luna. See you next time.