I was on a call last week with a founder who had 100,000 Discord members, a token that had launched three months earlier, and roughly 37 people who seemed to care. He knew the numbers were bad. What he couldn’t figure out was why. His co-founder had spent six months building the server: giveaways, role channels, a verification bot, a roadmap channel with emoji polls. It looked like a community. It behaved like a graveyard with good lighting.
This is the shape of marketing in Web3 right now. The channel strategy that worked in 2021 of loading up on Discord members and airdrop until something sticks stopped working sometime around the time people realised most airdrops were just paid retention on a phantom user base. The a16z crypto team put it plainly: growth hacks and continuous airdrops disconnected from real strategy look less like a path to product-market fit and more like a substitute for admitting you haven’t found it.
And most teams don’t have a marketing department. They have a founder who posts irregularly on X and a part-time community manager who is burning out. So the question is not how to build a marketing machine. It is how to do marketing that works with no team at all.
Why borrowed voice doesn’t work
The most common response to having no marketing team is to hire someone to talk for you. A PR agency, a ghostwriter, a social media manager who has never used the product. The Variant Fund team has watched this fail repeatedly: they point out that many founders know they need to step into the light, but feel blocked, so they hire speakers to speak on their behalf, and it rarely works because borrowed voice doesn’t build trust.
I have seen this play out more times than I can count. The posts are technically fine. The grammar is correct. The structure is reasonable. And no one cares, because the person writing them does not have the conviction that comes from having built the thing. Trust in Web3 is already battered. When a founder communicates through a filter, the audience feels the distance. It reads as marketing in the worst sense: a thing that was produced rather than a thing that was said.
Most technical founders have never practised it. But Variant makes the point plainly: in a market where attention is scarce, silence is not neutral. Showing up publicly is part of the CEO job, not a vanity project.
How to spend money when you have almost none
Let us assume you have a small budget. Not zero, but not enough for a retainer agency or a KOL campaign. The HypeLab team ran the numbers on minimum viable ad spend in 2026: a $5,000 test budget on a crypto ad network delivers approximately 2 million impressions and 16,000 clicks. Not nothing: enough to test a message, a landing page, and a funnel in a weekend.
The question is where to put those impressions. The old Web2 playbook says buy social ads and track conversions through third-party cookies. But third-party cookies are blocked by Safari, Firefox, and Brave, which collectively serve over 100 million users. Tracking pixels are blocked by privacy browsers and ad blockers covering 42% of users, per HypeLab’s analysis. The Web2 playbook fails not because crypto is unique but because the targeting infrastructure Web2 marketers rely on has been dismantled.
Display ads on crypto-native networks, by contrast, run at $0.10 to $2.00 CPC, according to HypeLab’s comparison of display versus KOL spend. The crypto advertising market exceeded $1.5 billion in 2026, with display advertising representing approximately 40% of total spend, per the same research on display ads in the marketing mix. That is a real channel, not a fringe experiment.
But the better investment for a team with no marketing function is not ad inventory. It is distribution through content that earns its way into places ads cannot reach. The Spindl team frames it well: crypto ads, when done well, are not just banners. They are relevant, native content inside an existing UI. Users engage with ads more than they do with non-sponsored content, which is a wild claim until you spend time in crypto-native platforms where the line between content and promotion barely exists.
The one thing founders assume that is wrong
The assumption that causes the most damage is this: big numbers mean big interest.
A founder sees 100,000 Discord members and feels good about it. But HypeLab’s analysis of vanity metrics across crypto communities shows that 100,000 Discord members means nothing if only 2% are active. A 100,000-person server often has less than 5% daily active users. The gap between total members and active members is where most Web3 marketing budgets go to die.
I have watched founders spend thousands of dollars on Discord growth campaigns that delivered exactly what they paid for: bodies in a server who never read a single message. The metric that matters is not membership count. It is whether anyone shows up when something important happens. A test I use with teams: post a substantive technical update and see how many people reply with something other than “gm.” The number is usually sobering.
The wrong assumption leads to the wrong spending. Budgets get allocated to volume tactics, server boost campaigns, airdrop farming, vanity KOL posts, that inflate the top of the funnel while doing nothing for retention or conversion. And because the numbers look big on a dashboard, the founder keeps optimising the wrong variable.
What a solo founder can actually do
The pragmatic path for a Web3 team with no marketing function has three layers, and none of them require a department.
Layer one: build in stages, not in silence.
Variant’s token launch checklist recommends building narrative momentum in stages: first tease the vision, then educate, then announce mechanics, and finally publish a participation guide. The instinct of most founders is to do everything at once — a website, a whitepaper, a token, a community — and then ask why no one is paying attention. Momentum doesn’t arrive on launch day. It builds across a sequence of smaller moments, each of which has one job.
Layer two: invest in the people who actually build.
The Talent Protocol team has been tracking a problem across the ecosystem: Base alone tracks over 40,000 builders, but only about 500 are rewarded in a given month. Grants, bounties, and incentive programs can attract builders, but they just as easily attract people optimised for short-term rewards rather than long-term commitment. The Hyperlane DevRel lead put it bluntly: the biggest issue is attracting quick-money builders who show up for the payout and leave before anything ships.
Retention of contributors is a marketing function. No one talks about it that way, but a project with five committed developers who ship every week will outperform a project with fifty grant farmers who disappear after the first milestone. Keeping the right people engaged is marketing, even if no one frames it that way.
Layer three: use incentives that are actually incentives.
The Galxe team published a case study on daGama, a project that ran seasonal growth campaigns across the platform. The results: 130,000-plus participations, 8,177 participants, and 4,370 contributors who generated 315 million impressions and 5.82 million engagements, with $115,000 in tokens distributed, 20% unlocked immediately and 80% over six months. The structure matters more than the amount. The vesting schedule separated the committed from the extractive. That is the difference between an incentive and a bribe.
Where the puck is going
The channels are shifting beneath us. Electric Capital tracks the macro numbers: prediction markets hit $44 billion in volume in 2025, up five times from the prior year. Stablecoins grew from $3 billion in 2019 to over $300 billion today. The user base is evolving, and the marketing tactics need to evolve with it.
The Multicoin Capital thesis on onboarding is worth sitting with. They argue there are only two ways to onboard to crypto: you buy in or you earn in. Over the past decade, people largely followed the first route. In the next decade, the majority will be through the second. That changes marketing fundamentally. If people are entering the ecosystem through work rather than speculation, the marketing message shifts from “get rich” to “get paid.” That is a different register, a different audience, and a different set of channels.
And the a16z crypto team is watching something stranger: AI agents becoming economic actors. Autonomous participants calling APIs, deploying capital, and executing transactions at machine speed. If a meaningful portion of on-chain activity in the next two years comes from agents rather than humans, then marketing to humans is only half the problem. You also need to be discoverable by software. That is a challenge that no amount of Discord growth will solve.
What a no-team marketing strategy looks like in practice
If I were advising that founder with the 100,000-member graveyard, I would tell him to close the server and start over. Not because community is unimportant, but because the server was a liability dressed as an asset. A small, loud, useful community of 500 people who actually use the product is more valuable for growth than a bloated server full of people who joined for a giveaway and never returned.
The budget would go to three things: a small display campaign on a crypto-native network to test positioning, a structured build-in-public cadence from the founders, and a contributor programme that rewards shipping, not joining. Everything else is noise.
The teams that will win in this cycle are not the ones with the biggest budgets or the biggest Discord counts. They are the ones that understood what Variant was saying about silence and did something about it. They showed up, wrote their own copy, defended their own decisions, and built a community of people who actually used the product. Not a marketing strategy, in the formal sense. Just what happens when you stop pretending you can outsource trust.
See also:
Leave a Reply