Hiring a Web3 agency is one of the highest-stakes decisions you'll make for your protocol. The gap between a shop that ships production-grade code and one that delivers a polished demo before disappearing is enormous — and it's rarely visible from a portfolio page or a sales call.
Here are the questions to ask before you sign anything, and the answers that should give you pause.
Why Vetting Matters More in Web3 Than in Traditional Software
Web2 mistakes are fixable. A broken API endpoint gets patched. A slow query gets optimized. In Web3, a vulnerability in a deployed smart contract can drain a protocol in minutes — and there's no rollback button on-chain.
That asymmetry means the agency you hire needs to be thinking about security, architecture, and chain-specific behavior from day one, not as a pre-launch checkbox. A shop that treats blockchain development as "just another backend" isn't a partner. It's a liability.
The Questions That Actually Reveal What an Agency Is Made Of
"Walk me through your audit process before mainnet deployment."
A serious agency has a real answer. They should describe internal review stages, the tools they use — static analysis, fuzzing, manual review — and whether they bring in third-party auditors for independent verification. They should also be able to tell you how past audit findings shaped their architecture decisions.
If the answer is vague — "we do thorough testing" — that's a problem. Testing and auditing are not the same thing.
"Which chains have you deployed to production, and what specific constraints did you have to work around?"
This is a depth probe. Anyone can list Ethereum, Solana, and a handful of EVM chains on a capabilities page. What separates real experience from marketing copy is the ability to describe concrete, chain-specific challenges.
Solana's account model requires a fundamentally different mental model than EVM state storage. Sui and Aptos use Move, which has ownership semantics that affect how you design assets and permissions. An agency that has genuinely shipped on these chains will have stories — edge cases, gas optimization decisions, protocol-specific gotchas. If they can only speak in generalities, they've likely done proof-of-concept work, not production deployments.
"Show me a project where something went wrong and how you handled it."
This one makes people uncomfortable, which is exactly the point. Agencies that have shipped real products have dealt with real problems: testnet behavior that didn't replicate on mainnet, a third-party oracle that behaved unexpectedly, a gas spike that broke user flows.
An agency claiming a spotless record either hasn't shipped enough to encounter problems, or isn't being straight with you. What you want is a clear account of what happened, how they diagnosed it, and what changed as a result.
"How do you handle scope changes mid-engagement?"
Web3 projects evolve. Token economics shift, a new chain becomes relevant, a regulatory consideration forces an architecture change. An agency with a rigid process that can't absorb that kind of movement will either deliver something outdated or charge you unpredictably for every deviation.
Ask for a specific example of a mid-project pivot and how it was managed. Look for process clarity: change request documentation, revised timelines, transparent repricing. Vagueness here tends to mean scope creep becomes a billing dispute later.
"What does your data infrastructure and indexing setup look like for a DeFi protocol?"
Frontend performance in DeFi depends heavily on how on-chain data is indexed and served. A shop that builds the smart contracts but leaves you to figure out your own subgraph, RPC infrastructure, or event indexing is delivering half a product.
Strong agencies think about the full data pipeline — how events are captured, how historical data is queryable, how the frontend stays in sync with on-chain state. If this question gets a blank look, you'll be managing infrastructure problems after launch.
"Who specifically will work on our project, and what's their background?"
Agencies sometimes sell you on senior engineers and deliver junior contractors. Ask for the names and backgrounds of the people who will actually write your code and conduct your reviews. Ask whether they're full-time employees or subcontractors. Ask what else they're running concurrently.
This isn't about distrust — it's about understanding what you're actually buying.
Red Flags to Watch For in Proposals and Conversations
Overpromising on timelines. A DeFi protocol with a custom AMM, liquidity management, and a full frontend does not get built in six weeks. Agencies that promise unrealistic timelines are either planning to cut corners or setting you up for a painful renegotiation.
No mention of security until you bring it up. Security should be woven into how an agency talks about architecture from the first conversation. If you have to ask about audits, that tells you where security sits in their priorities.
Generic portfolios without technical depth. UI screenshots and token names aren't a portfolio. If there are no technical writeups, no architecture diagrams, no explanation of what was actually built — that's a surface-level presentation. Ask for case studies with engineering detail.
Inability to discuss your specific chain. If you're building on Arbitrum and the agency can't speak to how its sequencer model or fee structure affects your design, that's a gap. Chain-specific knowledge matters, whether you're targeting Solana, a Move-based chain, or an EVM L2.
Pressure to decide quickly. Legitimate agencies have pipelines. They don't need to manufacture urgency. High-pressure tactics in a technical services context are a signal that something else is going on.
What a Serious Agency Looks Like in Practice
A genuinely capable Web3 agency covers the full product lifecycle — not just the parts that are easy to demo. That means architecture decisions informed by security modeling, smart contract development with audit preparation built in, data infrastructure that supports your frontend and analytics, and cloud operations that keep things running after launch.
When you're evaluating shops, look for evidence of that full-stack thinking. Do they ask about your indexing needs before you bring it up? Do they have a position on which audit approach fits your contract complexity? Are they thinking about your operational environment, not just your code?
DappWeb is built around exactly this model — engineering across Solana, EVM chains, Sui, and Aptos, with security, data infrastructure, and cloud operations treated as part of the same engagement, not separate line items.
Chain-Specific Questions Worth Adding
If your project is on a specific chain, add targeted questions to your vetting process.
For EVM work, ask about Ethereum smart contract development patterns they've used in production and how they approach upgradability and access control.
For Arbitrum-based DeFi, ask how they've handled Arbitrum DeFi frontend and analytics development — specifically how they manage latency between L2 state and frontend display.
For account-based chains like NEAR, ask how they've approached NEAR DApp development and account workflow design, particularly around named accounts, access keys, and contract-level permissions.
These aren't trick questions. They're baseline checks for whether the agency has actually shipped on the chain you care about.
FAQs
What is the most important thing to verify before hiring a Web3 agency? Verify that they have shipped production deployments on the specific chain you're building on — not just testnet or proof-of-concept work. Ask for technical case studies, not portfolio screenshots.
How do I know if a Web3 agency's smart contract audit process is credible? Ask them to describe their internal review stages and whether they use third-party auditors for independent verification. A credible process includes static analysis, manual review, and documented findings — not just "thorough testing."
Should a Web3 agency handle data infrastructure, or is that separate? A full-service agency should address data infrastructure as part of the engagement. On-chain data indexing, subgraph setup, and RPC infrastructure directly affect your frontend performance and analytics. Treating it as a separate concern usually creates post-launch operational problems.
What does it mean when a Web3 agency can't speak to chain-specific constraints? It typically means their experience is surface-level or limited to a different chain. Each major blockchain has architectural differences that affect how you design contracts, manage state, and handle fees. Agencies without that chain-specific knowledge will make costly assumptions.
How should a Web3 agency handle scope changes during a project? They should have a documented process: a formal change request, a revised timeline, and transparent repricing. If they can't describe how they've handled a mid-project pivot before, scope changes are likely to become disputes.
Is it a red flag if a Web3 agency doesn't mention security upfront? Yes. Security should come up naturally when an agency discusses architecture — not only after you prompt them. If audits and threat modeling aren't part of the first conversation, security is reactive in their process rather than foundational.
What's the difference between a Web3 agency and a smart contract auditing firm? An auditing firm reviews existing code for vulnerabilities. A Web3 agency builds the product from architecture through deployment. The best agencies incorporate audit-quality thinking throughout the build — and some bring in external auditors as an additional layer before mainnet.
The right Web3 agency asks hard questions before you do. They push back on timelines that don't account for audit cycles, flag architectural decisions that create security surface area, and treat your post-launch operations as their problem too. If the agency you're evaluating isn't doing those things, keep looking.