AllianceDAO: Can doing "non-scalable" things work in crypto?
TechFlow Selected TechFlow Selected
AllianceDAO: Can doing "non-scalable" things work in crypto?
Some of the most successful crypto startups have gone to extraordinary lengths to acquire and delight their users, whether those users are consumers or enterprises, and regardless of whether they have a token.
Navigating Web3 tides with focused insightsWritten by: Qiao Wang
Compiled by: TechFlow
Exactly ten years ago, Paul Graham wrote an article titled Do Things That Don't Scale. In my opinion, this is the most important piece Y Combinator (YC) has ever published for the startup community.
The core idea is that startups in their early stages must 1) manually recruit users and 2) manually deliver exceptional experiences to them. These are things almost no large company does, which is why they're considered "not scalable."
Since launching AllianceDAO three years ago, I’ve been reflecting on how well this advice applies to crypto startups. Over the past three years, I’ve accumulated enough data to answer this question.
But before I answer it, let me introduce you to some of the “not-scalable” things AllianceDAO alumni have done to attract and delight their users.
Synthetix (ALL1)
Synthetix was one of the earliest derivatives trading platforms and has consistently ranked among the top 100 protocols by FDV.
Kain recently returned to give a talk to ALL11 founders. He said something that would surprise most people: “You have to play the short-term narrative game.”
When Synthetix started, at the beginning of DeFi Summer in 2020, the prevailing narratives in the DeFi community were “liquidity mining,” “token staking,” and “TVL.” By 2023, these concepts had largely fallen out of favor. In fact, there's now a consensus that liquidity mining, for example, is detrimental for early-stage projects.
Yet in 2020, Kain embraced all of them. They ran liquidity mining programs, implemented token staking, and constantly talked about TVL on Discord and Twitter. And it worked. They attracted influential figures like DegenSpartan, who became passionate advocates.
Now, I'm not saying strategies like liquidity mining still work today. But the advice to embrace short-term narratives remains valid—especially when building a crypto-native product. The reason is simple: the crypto-native community is still relatively small, with everyone only one or two degrees apart.
Additionally, Kain was one of the most active members in his own Discord, frequently engaging with the Synthetix community. To this day, he regularly publishes blog posts about Synthetix—clearly aimed at attracting more users.
0x/Matcha (ALL1)
0x was one of the earliest DEX protocols. They also built Matcha, a leading DEX aggregator.
0x was ahead of its time. When it first launched, professional market makers were skeptical of DeFi and DEXs. It was nearly impossible to get them to provide liquidity on 0x.
So the 0x team built their own internal market-making infrastructure (later spun off as Periscope Trading). This served two purposes: first, to bootstrap liquidity—since professional market makers only care about exchanges that already have significant volume; second, to experience firsthand the pain points of providing liquidity, which helped improve the user onboarding process. Once organic liquidity grew sufficiently, they shut down their proprietary system.
Ribbon/Aevo (ALL2)
Ribbon was the first and most active structured products marketplace. They’re now building Aevo, a leading options protocol.
Julian also came back to speak at ALL10. Like nearly every successful DeFi protocol, he personally reached out to the first users. Interestingly, he noticed the same 20 people appeared across every options protocol’s Discord, so he messaged each of them and brought them into the Aevo community—because they were already naturally interested. The DeFi derivatives community is still niche enough that you can actually talk to everyone.
But he did more. Because on-chain options liquidity was generally limited, some Aevo users couldn’t find sufficient entry or exit liquidity. Julian manually sourced liquidity for them and posted bids/offers directly on their order books.
Mux (ALL3)
Mux is a top-three perpetual swap trading protocol on Arbitrum.
Mux (then called Mcdex) launched on Twitter in 2020. DeFi Pulse and Bankless stumbled upon Mcdex, liked the product, and reached out to Jean and Jie. The latter held separate calls with DeFi Pulse and Bankless, negotiating deals where both would write about Mcdex in exchange for referral fees. The product took off from there.
More importantly, Jie never stopped pushing. He executed what every Chinese startup does—the standard 996 grind: working six days a week, from 9 AM to 9 PM. I hear it’s evolved into 997 now. I often lurk in their Telegram group and see him replying to users at 11 PM local time. Remember, he’s a technical founder. I rarely see technical founders do this.
Dodo (ALL3)
Dodo is also one of the leading DEXs, holding a 5% market share.
Dai Dai worked at DDEX in 2018, another early DeFi protocol—two years before founding Dodo. In her spare time, she created a blog called “DeFi Labs,” where she produced educational content about DeFi.
Then, she realized she wanted to personally know her readers. So she pulled them into a WeChat group called “DeFi the World” and regularly organized offline meetups with members. This became China’s first—and largest—DeFi community. Let’s be clear: this was during the last bear market, when DeFi wasn’t really a thing yet, so “largest” meant around 100 people.
When she launched Dodo in 2020, those 100 people became its first 100 users.
Pendle (ALL4)
Pendle is currently the largest yield trading protocol.
I clearly remember TN inviting me to speak at their internal all-hands meeting to boost team morale. This was not scalable for either Pendle or me—but once I received the invitation, I thought to myself: “TN is special.”
But he went much further. Early on, he contacted potential LPs (liquidity providers) and scheduled one-on-one meetings to pitch Pendle’s yield opportunities. Before Pendle, he had no sales experience, but after enough attempts, he got better at pitching and started seeing conversions. Most founders don’t realize they can become great salespeople—sales is a numbers game. He also occasionally shared yield-trading ideas with potential LPs. As they became more familiar with the team and protocol, many converted into real users.
Like Synthetix, he also identified key influencers he wanted to collaborate with and reached out via cold DMs or emails. Today, most decent founders do this—but TN went the extra mile and did his homework. For example, if he wanted to partner with a specific influencer, he’d study their interests and habits, then draft materials he believed the influencer could publicly use. He kept doing this until he saw more organic mentions on social media.
Charmverse (ALL7)
Charmverse is the leading web3 community operations platform. (It was competitive two years ago, but most competitors collapsed during the bear market.)
Charmverse’s first version was a token-gated Notion. But obviously, building a Notion clone is extremely hard. So Alex and Matt hacked into Notion’s internal API and offered token gating as admin Notion users. (At the time, Notion hadn’t released its public API—now it has.) That’s how they landed their first DAO customers.
Of course, Alex attended many crypto conferences, building relationships with key decision-makers in DAOs—many of whom later drafted proposals to adopt Charmverse, which were eventually voted in by the DAO.
One particularly labor-intensive effort Alex made to convert existing Notion users to Charmverse was manually migrating their Notion content to Charmverse, followed by onboarding sessions teaching DAO members how to use the platform.
Hubble (ALL7)
The Hubble team built Kamino, the number-one yield vault marketplace on Solana.
When FTX collapsed in November 2022, everyone (except me) told Marius and Thomas to abandon Solana and migrate to Ethereum L2s. This thought crossed every Solana builder’s mind.
I remember discussing this weekly with Marius and Thomas for months. Ultimately, they concluded staying on Solana was the right move because:
-
Solana had soul: there were still developers passionately committed to it.
-
The tech was actually solid—it was fast and cheap. The downtime issues were engineering problems that could be solved, and they ultimately were.
-
Anatoly is the most inspiring L1/L2 leader after Vitalik.
Meanwhile, if they migrated to Ethereum, it was unclear how they could compete with established projects.
Marius and Thomas made a very difficult but ultimately wise decision: stick with a smaller, growing market they deeply understood (Solana), rather than jumping into a larger, more scalable one where they had no unfair advantage (Ethereum).
Synquote (ALL7)
By trading volume, Synquote is the number-one decentralized options trading protocol.
When Synquote launched, Ahmed ensured his team was available 24/7 to respond to user requests. One morning, a major whale wanted to execute a large trade. The team built a makeshift OTC product within hours, provided liquidity, and completed a record-breaking $3+ million notional trade that afternoon.
Providing 24/7 support led to an unexpected benefit: users who witnessed the product rapidly improving through direct engagement—and met the team—became passionate advocates.
Tensor (ALL8)
Tensor is the number-one NFT marketplace on Solana.
Ilja and Richard used an on-chain tracking tool to identify all major Solana NFT whales, found their Twitter handles, and DM’d some of them directly.
But that wasn’t enough. At the time, they were competing against NFT marketplaces like Magic Eden with orders of magnitude more distribution. So they formed alliances with other NFT marketplaces. Notably, they struck a deal with Hadeswap—which had 10x more distribution than Tensor—where Tensor would build a brand-new frontend for Hadeswap in exchange for backlinks on Hadeswap’s site. This dramatically increased Tensor’s visibility.
Of course, Tensor has made many other excellent decisions in its short history, but the partnership with Hadeswap was a pivotal moment.
Liquifi (ALL8)
Liquifi is the top token vesting solution for crypto protocols—think of them as the crypto version of Carta.
My office hours with Robin and Oliver were among my most enjoyable. They came to every meeting fully prepared. They’d show me a list of other AllianceDAO startups and ask me one by one whether I thought they were good prospects, then follow up. Incidentally, several AllianceDAO startups acquired their first customers through our community.
Robin and Oliver also scoured all public fundraising data on Twitter, Messari, and Crunchbase to identify potential clients. They asked every investor for intros to decision-makers. They even entered random Discord channels, directly asking people for referrals.
Robin described Liquifi’s customer support approach as “over-communicative.” For example, the team would review spreadsheets of vesting schedules sent by clients to reduce onboarding friction. They caught numerous errors—once even spotting that someone had been underpaid for two years. Token vesting is a highly sensitive matter, and this level of diligence helped clients feel comfortable partnering with Liquifi.
Clique (ALL8)
Clique is the number-one issuer of off-chain credentials.
During the Alliance program, Jaden interviewed over 200 protocols to validate use cases for offline credentials. They found their first strong use case in the Lens Protocol: combating impersonation.
Jaden reached out to the Lens team. At the time, Jaden had just moved to New York—where the Lens team was based. After three weeks of back-and-forth, the Lens team finally agreed to an in-person meeting. On the day of the meeting, however, they informed him they’d flown to ETHSF and wouldn’t attend. Without a ticket, Jaden flew to San Francisco that same day, waited outside the event afterward, and eventually found Lens’s partnerships lead. He successfully pitched the deal and secured the contract.
Next, Jaden and Kevin built a series of ad-hoc features to meet Lens’s needs. Jaden half-jokingly told me, “Clique is the least scalable product in crypto.” But eventually, Optimism and other L2s noticed Lens was using Clique, reached out, and became customers. Things that don’t scale immediately might accidentally scale later.
Yakoa (ALL8)
Yakoa protects prominent Web2 and Web3 brands from on-chain IP infringement.
Graham’s B2B sales strategy is “show up in person.”
For one client, Graham and Andrew flew across the country twice to meet them during negotiations. They spent a full day working in the client’s office, getting to know the entire team and internal dynamics. They sat down with leaders across multiple levels—something nearly impossible to coordinate over phone calls—to gain their buy-in. They had dinner with the engineers who would implement the integration. Throughout this process, they also uncovered additional pain points for cross-selling.
Stride (ALL9)
Stride is the number-one liquid staking protocol on Cosmos.
When Stride first launched, Aidan, Vishal, and Riley took turns being on call 24/7 until they confirmed the protocol was secure. For the first two weeks, one of the three woke up at 3 AM every night to ensure regular IBC packets were correctly relayed to Cosmos and that protocol accounting matched.
While most founders are active in their own Discord and Telegram, Stride’s team deliberately monitored other Cosmos community servers (like Osmosis and Mars) where Cosmos users hung out, answering questions. Since users across the Cosmos ecosystem could use Stride to liquid-stake their native tokens, this proactive support paid off. Eventually, they even set up automated jobs to crawl these Discords and dump relevant messages into their Slack for easier response, reducing the need for constant monitoring.
Primodium (ALL9)
Primodium is the most popular fully on-chain game by DAU (daily active users).
Emerson and Morris had some non-crypto background. With help from Will Robinson on our team, they built a solid network in the on-chain gaming community. Like Ribbon in the derivatives space, the on-chain gaming community is niche enough that you can reach most of the “important” players.
So Emerson and Morris posted the game on Twitter and introduced it to everyone in their network. Then, it spread privately, and eventually gained organic traction on social media. There was even a Wall Street Journal article mentioning it.
The lesson? When communities are small, you don’t need dramatic efforts to launch. Invest effort in building tight relationships. Those relationships will pay off from day one.
Sleepagotchi (ALL9)
Sleepagotchi is a sleep tracker with gamification elements and has the highest 30-day and 90-day retention I’ve seen among any crypto consumer app.
Anton viewed the Axie Infinity community as potential early adopters. At the time, “play2earn” was trending, so Sleepagotchi marketed itself as “sleep2earn.” Coincidentally, this ties back to Kain’s point about “playing short-term narratives.” Anyway, Anton offered Axie Twitter users a “free 3D avatar conversion” service. Users would share their favorite 2D Axies, and the team would create 3D versions. This caught the attention of Axie co-founder JiHo, who retweeted their work. JiHo later became Sleepagotchi’s angel investor after its June 2022 launch at NFT NYC.
Anton applied a similar approach with the BAYC community. Knowing Eminem owned a BAYC, he decided to create a 3D version of his NFT, showing it singing in front of a Sleepagotchi character. Fortunately, Eminem and Snoop Dogg dropped their song during NFT NYC in June 2022, and Anton released the 3D video the next day (as it was already ready). The tweet went viral, reaching 30,000 views within hours.
Crucially, Anton funneled potential users into their Discord instead of their mobile app—prioritizing community first. Joining Discord has lower friction than installing an app. Anton found many joined Discord just for fun, curious about the trendy “sleep2earn” concept. But for every question—serious or not—his team responded with 2–3 paragraphs. Those initially intrigued realized the team took the idea seriously and eventually downloaded the app.
Refract (ALL10)
Refract is the leading “crypto antivirus” for consumers.
When Refract first launched, the NFT platform Premint got hacked. Nish DM’d everyone on Twitter who said they’d been compromised—not with a pushy “use Refract,” but with “let’s understand what happened.”
Then, whenever a major crypto hack occurred, Nish wrote a post-mortem thread on Twitter. I’ve seen some of these threads—they had nearly unmatched engagement in crypto because people get emotional when losing money. Once, Nish and Justin showed me metrics revealing a striking correlation between spikes in new users and major hacks.
Caldera (ALL10)
Caldera is a leading Rollup-as-a-Service (RaaS) provider.
The most important thing Matt did to win early B2B customers was build small, temporary features for them—each taking 2 to 5 engineering days. One example was support for the secp256r1 curve (EIP7212). Yes, it’s so obscure most founders wouldn’t think it matters.
But such work generated immense goodwill. Customers were thrilled Caldera offered such high-touch, customized support and organically told their friends—leading to more deals. Remember, in crypto B2B, customers all know each other.
Interestingly, Jaden from Clique is Matt’s friend. Over the past two years, the most frequent thing Jaden told Matt was: “Are you sure this scales?”
Guardrail (ALL10)
Guardrail is the leading protocol security monitoring and investigation tool.
Their first customer happened to be another AllianceDAO alum: Pendle. Sam joined their Discord as a user, gave product feedback, connected with their dev-rel lead, then asked to be introduced to their internal tools lead—who eventually became Guardrail’s core user. I know this sounds confusing. But the point is, with B2B customers, it often takes time and multiple hops to reach internal champions and decision-makers.
Moreover, the customer (Pendle) was in Singapore, so most communication happened across a 12-hour time difference. Sam handled all customer support between midnight and 2 AM local time. After a demo, the customer requested unique features like auto-refresh so they could keep Guardrail running on screens all day. Sam and James built it for them the same week.
Kravata (ALL11)
Kravata is Latin America’s leading fiat on-ramp provider, serving some of the region’s largest exchanges.
Felipe understands that behind every tech company are humans. People enjoy eating, drinking, and talking. I find this especially true in countries with weaker rule of law—relationships matter more.
To land early clients, he met them in person, sharing coffee and delicious desserts. During the first coffee, he always insisted on paying—out of his own pocket, not the company’s. Like Yakoa’s team, he sometimes flew to meet clients. He discussed shared concerns, like regulations affecting both parties. Sometimes, mutual connections emerged—useful when building homophilic relationships, as people tend to connect with those similar to themselves. He offered business advice and helpful intros—no strings attached. If you’ve met Felipe, you’d agree he embodies humility and integrity, which certainly helps.
Critically, he did all this *before* direct sales—to make clients feel comfortable.
Do Things That Don’t Scale
After reading all these stories, I hope you arrive at a similar conclusion as I did. “Doing things that don’t scale” still broadly applies in crypto. Regardless of whether their users are consumers or enterprises, whether they have tokens or not, some of the most successful crypto startups have done extremely labor-intensive things to acquire and delight their users.
But there’s one caveat. Crypto does offer a natively scalable way to acquire early users: token incentives. Token incentives may be one of crypto’s biggest unlocks. Other major unlocks include asset ownership (Bitcoin), data ownership (crypto social), permissionless programmability (Ethereum), fundraising (ICO), permissionless financial services (DeFi), culture (NFTs), and tokenization of everything (RWA).
Token incentives have evolved significantly since Bitcoin—from instant rewards to retroactive airdrops to point-based non-token systems. As a rule of thumb, these are powerful, scalable user acquisition strategies worth studying deeply.
However, token incentives aren’t magic. They don’t replace the unscaleable work—they complement it. Many of the AllianceDAO alumni above ran token incentive programs *while* doing the heavy lifting.
Even for crypto protocols where tokens play such a central role they’re practically the product—such as DePIN, stablecoins, StepN (ALL7), games, and L1s—growth often began with founders leveraging friends, investors, and random contacts to kickstart momentum on Twitter.
In 2008, Satoshi exchanged numerous emails with fellow cypherpunks (including Hal Finney and Wei Dai), who then evangelized Bitcoin on bitcointalk.org and in person to early enthusiasts. The Ethereum founding team did a world tour in 2014, preaching to a small group of early developers—including the famous Devcon 0.
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
Alliance
