Owning projects
Dependencies and unblocking
Other teams will not feel your deadline. Ticket, follow up, escalate with options — and parallelize while you wait.
Other teams will not feel your deadline in their bones. The default state of a project is stuck. Your job is to notice stuckness within hours, not weeks, and treat dependencies as first-class work.
Walking through a blocked allowlist
- Name them early in the design doc and kickoff (team, API, SLA, named contact).
- Put asks in their tracker, not only in Slack. Tickets create a queue they already manage.
- Follow up on a schedule (for example every two days) with new information, not guilt.
- Escalate kindly and early. Escalation is "this unmet dependency will miss a date the business cares about." It is not an accusation. Copy your manager when the date is at risk so they can trade priority with their peer.
Give escalations options: "Approve allowlist by Friday; or we ship internal-only on date; or we move the external launch one week." Options make it easy to decide.
While you wait:
- Parallelize. FraudSvc is blocked? Finish permissions, docs, dashboards, or the next milestone's scaffolding.
- Substitute. Mock the dependency behind an interface; land the rest; swap when unblocked.
- Shrink. Cut the dependent feature from v1 if it is not the core bet. See MVP scoping.
- Unblock others. Answer review questions the same day; provide a minimal repro; pair for an hour instead of a week of async thrash.
Being the person who removes blockers for other people is how you become the DRI others want on the next project.
Midway: 40% of eng-days burned, an ACL spike revealed you need Identity's new API (ETA +3 weeks). Status red with three options: (A) ship folders without inheritance using flat ACLs by original date; (B) wait for Identity and move launch 4 weeks; (C) park the project. You recommend A, with a follow-up epic for inheritance. Leadership can choose. They cannot choose if you only say "it's hard." Do not disappear into nights and weekends hoping to invent a fourth option.
Common mistakes
- Waiting politely forever in Slack with no ticket and no date.
- Escalating as accusation instead of impact plus options.
- Sitting idle while blocked. Parallelize, mock, or cut.
Check yourself
You pinged Slack two days ago with no ticket. Next move?
File it in their tracker, follow up with the impact date, then escalate with options if still blocked.
Identity is +3 weeks and you have burned 40% of the eng-days. What do you bring to leadership?
Options with a recommendation — thinner MVP, moved date, or pause — not "it's hard."
Key takeaways
- Name dependencies early; ticket them; follow up on a cadence.
- Escalate with options and dates, not guilt.
- Parallelize, mock, or cut while you wait.
- Red with options is leadership's job. Heroic silence is yours, and it fails.
Further learning
Next: launch readiness.
Related
Driving projects to launch
Kickoffs, status, dependencies, unblocking, launch checklists, and a finish decision-makers recognize as shipped.
Green, yellow, red status
Status is a forecast of launch, not a diary of tasks. Yellow on Tuesday beats red the day before.
Collaboration and disagreement
Build trust before you need it, disagree with data and options, and make cross-team asks easy to grant.