Developer Activity
Who is winning the race to attract developers?
Developer activity reveals where ecosystems are heading. Network effects compound—more developers means more tools, more tools means more developers.
What To Track
| Metric | Signal |
|---|---|
| Monthly active developers | Ecosystem health |
| New developer retention | Onboarding quality |
| Full-time vs part-time | Commitment level |
| Cross-chain migration | Where momentum flows |
| Time to first contribution | Friction in the system |
Reports
Evergreen sources:
Historical snapshots:
- 2023 Jan | Electric Capital
- 2022 Oct | Alchemy
- 2022 Jan | Electric Capital
- 2020 Dec | Electric Capital
Web3 Workflow
Building on crypto requires navigating tradeoffs. A common progression:
- Hack with smart contracts and data inputs
- Use Scaffold-eth or generic components to prototype
- Build tests as concepts take shape
- Create the sales app with optimized UI/UX
- Review patterns, extract to library layer
- Fuzz testing and refinement
- Deploy to mainnet
Stack Decisions
| Layer | Choices |
|---|---|
| L1 blockchain | Ethereum, Solana, etc |
| L2 / Sidechains | Rollups, validiums |
| Build tools | Foundry, Hardhat |
| File storage | IPFS, Arweave |
| Chain querying | The Graph, custom indexers |
| Oracles | Chainlink, Pyth |
| Identity | ENS, Ceramic |
| Client libs | Viem, Ethers.js |
Key questions:
- How cost effective?
- How deep is lock-in?
- What are the architecture trade-offs?
Rollups-as-a-service like Conduit reduce infrastructure decisions.
Reducing Friction
- TypeScript end-to-end (API + frontend)
- Solidity + Foundry for contract testing
- Decision checklists for onchain vs offchain
- Don't rebuild what exists
Connection
- Developer Experience — The principles that make ecosystems attractive.
- Blockchain Decisions — Technical architecture choices.
- Predictions — Where we think developer activity is heading.
Links
Questions
Which metric type — leading indicator, lagging indicator, or health signal — is most commonly confused with the others in business performance analysis?
- At what measurement frequency does tracking this metric produce anxiety that interferes with the behavior it's designed to improve?
- How do you distinguish a metric that's genuinely predictive of outcomes from one that's simply correlated with other metrics you already track?
- Which metric would you stop tracking if you could only keep five — and does that reveal which metrics are actually load-bearing?