The real cost of an in-house team
The spreadsheet most founders build stops at salary. A realistic model doesn't. A senior Web3 marketing lead runs $140k–$200k base; add a content/social hire ($70k–$100k), a community manager ($60k–$90k), and you're at $270k–$390k in payroll before benefits, tokens, or equity.
Then tooling: analytics, social scheduling, email, design, KOL databases — $2k–$5k monthly. Then the hidden line: hiring takes 2–3 months, ramp takes another 2–3, and a mis-hire at the lead level costs you six months and roughly a full salary in sunk cost.
Fully loaded, a minimum viable in-house team is $350k–$500k in year one, with meaningful output starting around month four.
The real cost of an agency
Crypto marketing agencies charge anywhere from $3k to $50k+ per month depending on scope and prestige. A boutique like Chalk Labs starts around $3k/month for a focused scope — PR, SEO/AEO/GEO, or performance marketing — scaling as channels stack.
What you're buying is not just hours. It's pattern recognition from dozens of launches, existing journalist and KOL relationships, and systems already built. An agency's first campaign ships in weeks because the machinery exists; your in-house hire spends month one buying tools.
The honest downside: agencies juggle multiple clients, will never absorb your culture the way an employee does, and knowledge walks out the door when the contract ends unless you deliberately structure handovers.
What agencies genuinely cannot do
Community is the clearest case for in-house. A Discord or Telegram community smells outsourced moderation instantly, and community trust is your most defensible Web3 asset. Founder-voice content is second — ghostwritten founder threads work, but the strategy behind them needs someone living inside the company.
Deep product marketing is third. An agency can position your protocol; it cannot attend every standup and catch the roadmap shift that changes the narrative.
This is why the binary framing fails. The teams that win pair an internal community/product-marketing core with agency leverage on specialized, spiky workloads: launch PR, SEO infrastructure, paid experiments, automation builds.
A decision framework by stage
Pre-seed to seed: agency-first. You cannot afford a great full-time lead, and a mediocre one is worse than none. Retain a boutique for launch-critical work and have a founder own community personally.
Series A: hybrid. Hire one strong internal marketer or community lead; keep the agency for PR, SEO/GEO, and performance — channels where senior specialists beat generalists.
Series B and beyond: in-house core, agency edges. Build the team, keep agencies for surge capacity (token launches, market expansions) and disciplines that are hard to hire, like GEO or growth engineering.
The sequencing mistake to avoid: hiring a junior in-house marketer first and expecting agency-grade output across six channels.
Questions we hear about this
Below roughly $120k/year of marketing spend, almost always. A $3k–$10k monthly retainer buys senior multi-channel execution that would cost $350k+ to replicate in-house. Above Series B scale, in-house economics start to win for core channels.
Community management and founder voice. Crypto communities detect outsourced moderation quickly, and community trust is the most defensible asset a Web3 project has. Keep those internal; outsource specialized channels like PR, SEO/GEO, and paid.
A competent agency ships first campaigns within 2–4 weeks because tooling, relationships, and playbooks already exist. An in-house hire typically needs 3–6 months from job posting to meaningful output.
Yes — that's the recommended sequence. Start agency-first pre-raise, hire an internal community or marketing lead at Series A, and structure the agency contract with documentation and handover clauses so knowledge transfers instead of walking out.