Treat Token Emissions as Marketing Spend, Not Policy

Most teams still talk about “emissions” like they’re running a central bank. They debate inflation schedules, runway, and “sustainable yield” as if they’re setting macro policy for a small country.

But at idea/early stage, your token is not a currency and your emissions are not monetary policy.

They’re coupons. Discounts. Bounties.

In plain terms: they’re marketing spend.

This piece argues you should plan, budget, and measure emissions the same way a performance marketer treats paid acquisition: with CAC, cohorts, and explicit stop conditions. If you can’t say what a dollar of token incentives is buying you — and when you’ll turn it off — you’re not “bootstrapping the network”.

You’re just paying people to dump your equity.

The “emissions as yield” meme comes straight out of 2020–2021 DeFi. Compound, Sushiswap, Curve, and others sprayed huge token rewards at LPs and lenders. On paper, it looked like 50–200% APY “yield.” In practice, most of that “yield” was just the protocol dropping more of its own token on you — a marketing rebate, not real, repeatable cash flow.

The trap was that both teams and users started treating these emissions like a base interest rate: a permanent feature that defined whether a protocol was competitive. From there you get architectures where emissions are hard-coded for 4–8 years, even if they’ve stopped attracting real users and are only feeding mercenary capital.

If you think of emissions as yield, you feel pressure to keep the music playing. If you think of them as coupons, you’re fine shutting the campaign down the moment it stops doing real work.

If your token represents a claim on future upside, emissions are simply a way to pre-allocate some of that upside to incentivize specific behavior today. It’s the same logic as retail coupons: “20% off your first order if you try us now.” No one expects the discount to be permanent, and no one mistakes it for the store’s long-term pricing. Your token should be treated the same way.

When you emit tokens to LPs, traders, or creators, you’re not defining a monetary policy. You’re running a paid acquisition campaign to buy liquidity, volume, or content. So you should be thinking like a performance marketer evaluating Facebook or Google ad spend:

• What exact behavior am I paying for?
• What is my cost per unit of that behavior?
• Over what horizon do I expect that behavior to become self-sustaining so I can shut the spend off?

If you can’t answer those, you’re not ready to turn emissions on.

If you want emissions to function as real marketing spend, you have to put them in dollar terms and connect them to specific behaviors.

First, price your token honestly. Use a conservative moving average of the secondary market price or, if you’re pre-listing, a steep discount to your last round’s implied valuation.

Second, define the exact behavior you’re paying for: “provide $1 of liquidity for 30 days,” “publish one piece of content that reaches X engagement,” “onboard one new paying user,” etc.

Third, calculate CAC. If you emit 10,000 tokens in a month and your honest price is $0.20, you’ve spent $2,000. If that produces 200 net-new users who remain active after rewards stop, your CAC is $10. If it produces $1m of TVL that disappears the moment incentives end, your CAC is effectively infinite.

The core metric is not APY; it’s cost per retained unit of behavior.

Paid acquisition only works if you’re willing to pause or kill channels that don’t pull their weight. Emissions should be treated exactly the same way.

Before you launch anything, set hard, explicit stop conditions: a calendar date, a budget ceiling, and a minimum CAC / retention bar you’re not willing to go below. Make it concrete and visible from day one:

“We’re running a 90‑day liquidity mining program with a fixed pool of 1m tokens. We’ll only consider extending if at least 40% of liquidity sticks after rewards end and CAC per retained LP is below $X.”

Then honor it. No exceptions, no “one more epoch just to see.”

When the data shows you’re just underwriting farm‑and‑dump behavior, you don’t tweak around the edges—you shut emissions off or slash them aggressively. Some users will get loud about it. That’s expected. The users who stay when the subsidies disappear are your actual product–market cohort.

The failure mode is to treat emissions as an entitlement or a “community right,” instead of what they really are: a line item in your growth budget that needs to earn its keep.

We’ve watched liquidity mining play out from both angles.

On the effective side, early Curve and Aave used emissions as a deliberate tool. They pointed rewards at specific pools to bootstrap depth where it mattered, then gradually dialed incentives down as real usage, trading volume, and fee revenue took over. Rewards were a temporary bridge to sustainable liquidity, not the product itself.

On the other side, a wave of 2021 DeFi forks — Vampire Sushi clones, OHM forks, “ultra-high-APY” DEXs — carpet-bombed the market with 1,000%+ APY, with the entire strategy reduced to “make TVL go up.” It worked until it didn’t: capital rushed in for the yield, then disappeared the moment emissions slowed, leaving a dead chart and a broken token behind.

The gap between those outcomes wasn’t philosophy; it was execution discipline. Teams that treated emissions as a finite campaign tracked cohorts, retention, and fee growth, and cut or redirected spend the second it stopped driving durable behavior. Teams that treated emissions as a standing policy simply kept printing until the market forced a hard stop.

If you’re an idea-stage founder, your token is not a macro asset. It’s closer to a stack of coupons you can spend to buy very specific behaviors while you grind toward product–market fit. Treat emissions like performance marketing: denominate them in dollars, measure CAC and retention, and shut them off fast when they don’t return the spend.

The protocols that made it through the last cycle weren’t the ones shouting the highest APYs; they were the ones that treated their token as a finite budget, not a belief system.

So as you sketch your token model, pressure-test every line item: if this were pure USDC spend on Meta ads, would I still do it? If the answer is no, why are you comfortable paying for it with the most expensive asset you have—your own equity?