Product Collaboration Best Practices
Solana products fail slowly when engineering and product talk past each other: "ship Friday" vs "audit blocked." These practices align roadmaps, specs, and communication without turning engineers into ticket machines or PMs into accidental security reviewers.
How to Use This List
- Review in eng+product kickoff each quarter
- Reference in sprint planning when friction spikes
- New PM onboarding reading list item
A - Discovery & Specs
- No sprint commit without written tech spec or RFC stub. Even one-pager for small features.
- Product defines user outcome and success metric. Engineering defines on-chain state change.
- Non-goals section required. Prevents "while you're at it" CPI integrations.
- Tokenomics changes include authority model. Multisig vs governance vs immutable.
- Design review scheduled before estimate finalization. For any user fund touch.
B - Planning & Estimates
- Three-point estimates shared with PM. Optimistic / likely / pessimistic visible.
- Audit and deploy buffers outside feature estimate. Named in roadmap slide.
- 20% capacity for harden/platform visible on roadmap. Not secret eng work.
- Dependency graph shared internally. Client blocked on IDL is visible to PM.
- Re-estimate trigger defined. Spec change, audit Sev-High, failed spike.
C - Execution & Communication
- Weekly eng-product sync 30 min. Blockers, not status readout.
- Demo includes explorer link or wallet flow. Slides-only demos insufficient.
- Single incident commander approves external posts. PM and eng aligned on facts.
- Delay announcements include trade-off and next checkpoint. No silent slip.
- Auditor status in shared tracker. PM not surprised by firm escalation.
D - Trust & Culture
- Disagree-and-commit documented. PM owns user bet; tech lead owns safety floor.
- Near-misses discussed in retro without blame. Product included monthly.
- Engineers invited to customer/community calls periodically. Ground truth empathy.
- PM does not commit mainnet date without eng devnet-done gate.
- Wins include security and ops milestones in all-hands. Not only feature GIFs.
FAQs
PM technical depth required?
Enough to read specs and ask smart questions; security review stays with engineering.
Who owns roadmap?
PM prioritizes outcomes; tech lead injects harden/scale with veto on safety gates.
Conflict resolution?
ADR for technical fork; exec for date vs safety with written risk memo.
DAO product?
Governance forum for params; eng blog for implementation - PM bridges both.
Remote collaboration?
Written specs over verbal; record design reviews for async PM in other TZ.
Metrics shared?
DORA + user metrics in same QBR slide - connect delivery to outcomes.
Feature flags product role?
PM defines rollout cohort; eng implements on-chain + client guards.
When PM in incident bridge?
Comms lead role; not every engineer-PM pair - rotate training.
Outsource design?
PM still signs spec acceptance before eng build.
First 90 days new PM?
Read ADRs, attend incident drill, ship one small devnet feature with eng pair.
Related
- Requirements to Technical Specs - spec quality
- Stakeholder Communication - external alignment
- Leadership Best Practices - tech lead side
Stack versions: This page was written for Agave 4.1.1, Solana CLI 3.0.10, Anchor 0.32.1, anchor-lang 0.32.1, Rust 1.91.1, @solana/kit 7.0.0, Surfpool 0.12.0, and LiteSVM 0.6.x.