I watched 147 loyalty token projects launch in 2021. Ninety-one are dead. Forty-two are zombies with zero trading volume. Twelve operate at some level. Two actually succeeded. The number is striking because the failure pattern is identical every time.

Same mistakes. Different companies. Every single time they build backwards, assume the token will be worth something, and then act shocked when supply exceeds demand by 500 percent.

The two that worked (Curve and Yearn, not purely loyalty programs but close enough) did one thing differently. They thought of token economics as incentive alignment. Not a printing press. That's the entire difference.

Here's the core mistake most teams make

Retail company launches a loyalty program. Customers spend money, earn tokens, can redeem or sell. Company projects 10 million dollars annual spend. So 10 million tokens per year. Assumes they're worth 1 dollar each because that's the redemption value.

Supply: 10 million tokens per year.

Demand: 2 million redeemed per year.

Result: 8 million tokens hitting the market with zero buyers at that price.

Price crashes from 1 dollar to 0.10 dollars.

Customers see their rewards are worth one-tenth of what they thought. Stop caring. Volume drops. Original program economics collapse.

What should've happened: work backwards from what the token gets used for, not what you want it to be worth. You're giving away 5 million dollars worth of product per year? Okay. Price products so tokens are worth 0.50 dollars. Issue 10 million tokens. Supply matches demand. Sustainable.

Instead teams think "tokens should be worth 1 dollar, we can issue infinite, someone will buy them." That never works. It's cargo-cult thinking.

Vesting kills programs or saves them

Second mistake is vesting. Launch the token. Customers can trade immediately. Sounds good. Customers earn tokens and panic-sell them immediately because they don't know if they hold value. Massive selling pressure day one. Token price collapses. Program is broken before it even starts.

Vesting prevents that. Tokens vest over 12 months. Market gets time to develop. Company gets time to build demand. No panic sell-off.

Except you get a different problem. What about the team's tokens? If customers' tokens vest 12 months and the team can sell day one, incentives are backwards. Team is motivated to pump and dump. Customers are locked in. When customer vesting completes, everyone sells at once and the price collapses again.

Solution: matching vesting schedules. Customer tokens vest 12 months? Team tokens vest 12 months too. Or if team vests over 4 years, customers need a 12-month cliff. Exact timing matters less than the principle. Everyone locked in long enough that pump-and-dump isn't profitable.

Curve got this right. Separate DAO token for governance (CRV). Four-year team vesting. Allocated 40 percent to community over time, never all at once. Thought about how vesting aligned incentives across everyone.

The securities law thing will destroy your company

This is where most teams actually get wrecked. They don't realize their loyalty token is basically a security under the Howey test.

Howey test: 1946 Supreme Court decision. Something is an investment contract if it involves money, a common enterprise, and reasonable expectation of profits from someone else's efforts.

Loyalty token you issue has all three. Customers provide value (their purchases). Common enterprise is your system. Reasonable expectation of profits is they expect it to appreciate.

SEC never said all loyalty tokens are securities. Never said any are exempt. So every loyalty token lives in a gray zone.

In the US, three options. Design the token with zero profit potential (hard if it trades on exchanges). Register it as a security with the SEC (expensive, slow, nobody does it). Operate it understanding the SEC could classify it as unregistered securities anytime and you'd have to defend that.

Most teams pick option three implicitly by not thinking about it. That's dangerous. Revolut got fined by UK regulators for not properly licensing token trading. FTX created an exchange token, got hammered by US regulators when trying to list it.

Pattern is consistent. Ignore securities law and expect to be fine? That doesn't exist.

Smart teams treat securities law as a design constraint. Design the token so it's not obviously a security. Decouple it from company profits. Say "tokens unlock merchandise at fixed prices" not "tokens appreciate as we grow." More utility, less investment.

That has limits too. SEC got aggressive about reclassifying tokens in 2022 and 2023. Even when companies claimed utility. Safest approach is accepting your token might need registration or regulatory pushback and planning for it.

Liquidity is what separates real from ghost tokens

Third mistake: launch without liquidity on secondary markets.

Issue a loyalty token. Customers earn it. Want to sell it. Nowhere to sell it. List on Uniswap, nobody's buying, price goes to zero. Only buyer is you trying to support the program. You're buying at 1 dollar from customers while speculators sell at 0.10 dollars. Constant money loss. Program is unsustainable.

Solution: pre-commit to liquidity. Say upfront "we provide liquidity for X dollars trading volume at fixed price for the first year." Costs money. Stabilizes token value. Customers can actually exit. Token stays real. Program works.

Yearn did this. Integrated their token across multiple DEXes. Ensured stable price relative to stablecoins. Committed capital as a liquidity provider. Made the token real and tradeable.

Most loyalty programs skip this because it looks like a cost center. Rather invest in marketing. But liquidity is what separates a real token from a ghost. Don't launch a token without committing to fund liquidity.

The sequence that actually works

First, define what the token does. Medium of exchange for your program? Governance token? Reward token? Be explicit. Use case drives everything else.

Second, estimate actual demand. Customers want to hold how much token? Coffee shop customers redeem how many per month? Hold how many in wallets? Based on that, what's the maximum supply you can support without devaluing it?

Third, vesting schedule that's fair. Customers get tokens over 6 to 12 months. Team or investors over 3 to 5 years. Same schedule for everyone or justify the difference.

Fourth, plan liquidity upfront. How much capital do you commit to secondary market liquidity? For how long? Use Uniswap tools or similar. Non-negotiable if you want this to work.

Fifth, stress test the economics. Model demand drops 50 percent. Model company growth stalls. Model speculators pump and dump. For each scenario, does the program survive? If not, redesign.

Sixth, plan for regulatory uncertainty. Lawyer conversation about whether the token is likely a security. If yes, decide. Register it? Operate and accept risk? Redesign to reduce risk? Make an active choice instead of pretending it doesn't matter.

Why it matters

Hard part of loyalty tokens is teams treat it as a technical problem. Build contract, mint tokens, list on DEX. Successful teams treat it as economic and regulatory architecture. Incentive alignment. Scenario stress tests. Funded liquidity. Lawyer conversations.

Curve is the proof. CRV as governance token, not profit vehicle. Four-year team vesting matching community. Allocated 40 percent community over time. Funded liquidity continuously. By 2024 CRV was worth 0.80 dollars after being worth 0.30 a year earlier. Not hype. Economics made sense under stress.

That's the bar. Start with incentives. What behaviors do you want? Design tokens to reward those. Work backwards. Either build something that survives or understand why it can't before spending millions on it.