# The NFT and token launch checklist: what to get right before mint day

*By Roberto Lazar, founder of Dock30 · Published 2026-07-24 · Updated 2026-07-24 · 8 min read*

A practical pre-launch checklist for an NFT or token mint: chain choice, contract, mint site, metadata, allowlist, audit, and rough 2026 launch costs.

Before an NFT or token hits mainnet, seven pieces have to work at once: the chain you picked, an audited mint contract, a mint site that survives a traffic spike, metadata and images pinned somewhere permanent, a working allowlist, a reveal that does not leak, and a launch runbook you have actually rehearsed. The launches that fail almost never fail on exotic contract exploits. They fail on a mint site that falls over, on metadata that renders blank, and on a gas war nobody planned for. Get those unglamorous parts right and you have removed most of the risk.

Web3 launches are where Dock30 started. Our first project in 2021 was a Solana NFT mint, a landing page plus a mint page, and we have shipped hundreds of Web3 projects since with public five-star [reviews](/reviews) to show for it. This is the pre-mint checklist we run before we let a client announce a date.

## The moving parts, and which ones actually break

A launch is not one deliverable. It is a system, and each part has its own failure mode.

| Piece | What it does | How it fails |
|---|---|---|
| Chain | Where the token lives | Wrong choice forces a gas war or scares off buyers |
| Mint contract | Enforces supply, price, allowlist | Reverts, duplicate mints, unbounded loops |
| Mint site / dapp | Where buyers actually mint | Collapses under load, RPC rate limits, wallet bugs |
| Metadata + storage | Images and traits | Unpinned IPFS, blank renders, reveal leaks |
| Allowlist / presale | Controls who mints first | Merkle root mismatch, double claims |
| Audit | Independent code review | Skipped, or invalidated by a post-audit change |
| Marketing + runbook | Demand and coordination | Hype with no capacity to serve it |

In our experience the contract is the part teams worry about most and the part that breaks least, because it is the part that gets tested. The mint site and the metadata are where drops actually go wrong, because they get built last and rehearsed never.

## Pick the chain before you write a line of contract code

The chain decides your tooling, your wallet support, your metadata standard, and how a popular mint behaves under pressure. Change it late and you rewrite almost everything.

For high-volume consumer mints, Solana fees stay **below $0.001** in normal conditions, and compressed NFTs let you mint hundreds of thousands of tokens for a fraction of a cent using state compression, which is not economical on any Ethereum chain, per [MEXC's 2026 Solana versus L2 analysis](https://www.mexc.co/en-IN/learn/article/solana-vs-ethereum-l2s-2026-fundamental-analysis-tvl-revenue-stablecoin-metrics/1). Solana also isolates congestion with local fee markets after SIMD-0096, so a hyped mint no longer makes every other app on the chain expensive.

An Ethereum L2 like Base, Arbitrum, or Optimism runs at roughly one cent to fifty cents per transaction, cheaper than L1 but still far above Solana, and it buys you the deepest ERC-721 marketplace liquidity and the audience that comes with it. Ethereum L1 itself only makes sense for ultra-premium 1/1 work where a twenty-dollar mint fee is noise against the sale price. Most consumer collections we ship land on Solana or an L2, and the deciding question is usually audience, not fees.

## The contract: batch minting, allowlists, and supply logic

If you are on Ethereum or an L2, do not deploy a naive ERC-721. A standard single mint costs around 70,000 gas per token, while [ERC-721A](https://www.alchemy.com/blog/erc721-vs-erc721a-batch-minting-nfts) drops the marginal cost of each additional token in a batch to roughly **5,000 gas** by writing ownership once for a consecutive range. For a buyer minting five in one transaction, that is real money saved, and it is the standard Azuki popularized for exactly this reason.

For the allowlist, store a Merkle root on chain and verify a proof at mint time, rather than pushing thousands of addresses into contract storage. The root is one 32-byte value; each minter submits a proof their wallet earned off chain. It keeps gas sane and the list flexible up to the last minute, as the [2026 whitelist guide from DEXTools](https://www.dextools.io/tutorials/what-is-nft-whitelist-mint-pass-guide-2026) lays out. Two things we always test here: that a valid proof for the wrong address is rejected, and that no address can claim its allocation twice. Double-claim bugs are common and embarrassing.

The rest of the contract checklist is the same discipline any Solidity deserves: bounded loops, checked external calls, correct access control on mint and withdraw, and a reveal-safe token URI. That last one matters more than people expect.

## The mint site is where launches actually break

The contract is public and audited. The mint site is the thing you built in the last week, and it is the single point of failure between your marketing and your buyers. This is the part we build as custom software, not a template, because [custom development](/services/custom-development) is what separates a mint that holds under a spike from one that does not.

Three things go wrong on drop day. The site itself goes down under a traffic surge it was never load-tested for. The RPC endpoint rate-limits you at the worst possible moment because everyone is reading contract state at once. And wallet integration behaves differently across Phantom, MetaMask, mobile in-app browsers, and the hardware wallets your whales use. Rehearse all three. Run a public testnet mint with real wallets and real load, not a demo on your laptop. Put your own RPC behind the site with generous limits, and cache read state instead of hammering the node.

Gas wars deserve their own line. When demand exceeds supply on a public mint, buyers bid up gas and losers still pay for their failed transactions, which is a terrible experience that gets blamed on you, as [DonkeVerse's guide to preventing gas wars](https://medium.com/donkeverse/101-guide-prevent-gas-wars-and-failed-mints-b425137ffbee) explains in detail. The fixes are structural: a real allowlist with per-wallet caps, a Dutch auction or timed phases instead of one open free-for-all, and a chain where fees do not explode under load. Design the mint mechanics to avoid the war, do not just hope buyers behave.

## Metadata, storage, and the reveal bug

Almost no collection stores images on chain, because it is expensive and rarely worth it. The token holds a URI and the art lives off chain, which means the art can vanish if you host it carelessly. Pin metadata and images to IPFS through a service that keeps them pinned, or use Arweave for pay-once permanence. Then verify every hash actually resolves before you launch, because the classic failure is an NFT that shows a blank or a placeholder on the marketplace and a support queue full of people asking you to refresh metadata.

The reveal is its own trap. If you plan a delayed reveal, the pre-reveal metadata must not point at the final art, and the final folder must not be public or guessable before reveal time. We have watched collections get sniped because the real metadata sat at a predictable IPFS path while the contract still showed the placeholder. Freeze the reveal metadata, flip a single base URI at reveal, and test that the pre-reveal state genuinely hides everything.

## What a launch actually costs in 2026

Costs vary wildly with ambition, so treat these as engineering ranges, not a quote. A standalone mint contract runs roughly $3,000 to $10,000, and an independent audit $5,000 to $20,000 depending on complexity, per [Octal's 2026 smart contract cost guide](https://www.octalsoftware.com/blog/smart-contract-development-cost). Chain fees to actually deploy and mint are minor next to that. The real budget goes to the contract, the mint site, the audit, and marketing, in roughly that order of how often they get underfunded.

| Cost area | Rough 2026 range |
|---|---|
| Mint contract (standard standards) | $3,000 to $10,000 |
| Independent audit (if funds are held) | $5,000 to $20,000 |
| Mint site / dapp | Scope-dependent, built custom |
| Chain deployment + mint gas | Cents to low thousands |
| Marketing and community | The line most teams overspend |

Where we fit: our Web3 work is [fixed-scope from **EUR 350**](/services/blockchain-web3), with the exact price and delivery date in writing before we start and 30 days of free support after launch. A first NFT mint, landing page plus mint page, is close to how we began in 2021, so it is well-worn ground for us.

## Audit and rug-pull optics

An audit is not optional the moment your contract holds funds, controls a treasury, or carries custom mint, payment, or upgrade logic. Schedule it weeks before launch, not days, because auditors find things and remediation plus a re-check takes time. The full pre-mint security gauntlet we run, the twelve code and process checks, lives in our [smart contract security checklist](/blog/smart-contract-security-checklist), and it applies to a token mint as much as to a DeFi protocol.

There is also the trust side, which is not a code problem but will sink a launch just as fast. Buyers now read for rug-pull signals: an unverified contract, a mint function the team can call to inflate supply, withdraw rights on a single anonymous wallet, an unrenounced owner with god powers. Publish the verified source, put privileged keys behind a multisig, document what the team can and cannot do, and say it plainly before anyone asks. Optics are earned by making the contract boring and legible, not by promising you are honest.

We have been shipping Web3 launches since 2021, and this checklist is how we gate ours. If you have a mint on the calendar and want a team that has done this a few hundred times to build the contract and the mint site, or just to stress-test the plan you already have, book a [free 15-minute call](https://calendly.com/dock30/15min) or send the details through our [contact page](/contact). We will tell you straight what you need and what you can skip.

## Frequently asked questions

**What do I need before I mint an NFT collection?**

You need seven things working together: a chain, an audited mint contract, a mint site that can handle load, hosted metadata and images on IPFS or Arweave, an allowlist mechanism, a reveal plan, and a rehearsed launch runbook. Most botched drops fail on the site and the metadata, not the contract itself. Test the full path on a public testnet with real wallets before you announce a date.

**Should I launch on Solana or an Ethereum L2?**

Solana suits high-volume, low-price consumer mints because fees stay below a cent and compressed NFTs let you mint at a fraction of a cent. An Ethereum L2 like Base or Arbitrum suits collections that want deep marketplace liquidity and an ERC-721 audience, at fees of roughly one cent to fifty cents. Pick the chain before you write contract code, because it dictates your tooling, wallet support, and metadata standard.

**Why do NFT mints fail on drop day?**

The three usual causes are gas wars during oversubscribed public mints, a mint site or RPC endpoint that collapses under a traffic spike, and metadata that renders as blank or wrong because IPFS pins were never verified. Failed transactions still burn gas, so a bad mint costs your buyers real money and your project its reputation. A rehearsed load test and a rate-limited allowlist prevent most of it.

**How much does it cost to launch an NFT collection in 2026?**

A simple collection on well-trodden standards runs from a few thousand euros for the contract and mint site, while custom mechanics, an audit, and heavy marketing push a serious launch well into five figures. A standalone mint contract typically costs 3,000 to 10,000 dollars and an audit 5,000 to 20,000. Chain fees are minor next to engineering and marketing.

**Do I need an audit before an NFT mint?**

If the contract holds funds, controls a treasury, or has any custom mint, payment, or upgrade logic, yes, budget for an independent audit weeks ahead of launch. A stock ERC-721A mint with no custom value flows can often be defended with static analysis and an experienced review instead. Either way, freeze the code before the review and re-check any change made after it.

---

Written by Roberto Lazar, founder of Dock30. Book a call: https://dock30.com/contact