How do we do this if we’re stuck on Microsoft Teams and ServiceNow?

Great—use what you have. Create Teams channels with a clear naming convention, wire ServiceNow change windows into a shared channel, and pre-approve a program change model. The patterns don’t require Slack or fancy tooling; they require consistency and ownership.

We already have standups and PI planning. Why add more cadence?

PI planning is a zoomed-out view. The heartbeat aligns weekly decisions across security, ops, data, and app so you don’t get multi-week drift. The goal isn’t more meetings; it’s fewer, smaller surprises.

What’s the minimum viable metrics set to start?

Three: (1) deployment frequency per stream, (2) MTTR for incidents tied to the initiative, and (3) time-to-decision for ADRs. Add SLOs for the first user-facing slice as soon as you can page on it.

How do we avoid weaponizing metrics against teams?

Measure at the stream/program level, not individual teams. Use metrics to ask better questions in steering, not to hand out gold stars. Tie improvements to removing blockers—funding, sequencing, and clear ownership.

We’re mid-flight and behind. Can we introduce this without stopping the train?

Yes. Start with the heartbeat and the steering doc this week. Add ADR gating on the riskiest repos. Catalog top dependencies in Backstage. You can roll these in without a freeze; in fact, they tend to reduce further slippage.

Culture · Oct 23, 2025 · 8 minute read

The Program That Stalled Until We Fixed the Humans: Cross‑Functional Patterns That Actually Ship

You don't need another deck about collaboration. You need rituals, behaviors, and guardrails that force coordination in real life—across infra, app, data, security, and the suits.

Back to all posts

The Program That Stalled Until We Fixed the Humans: Cross‑Functional Patterns That Actually Ship

Key takeaways

Implementation checklist