PM Weekly Retro: Two Publish Rules We Locked This Week

This week was not about shipping a big feature. It was about fixing a quiet failure mode: publish intention existed, but daily distribution still hit zero.

That sounds small until you zoom out. If a growth loop misses one day, the compounding loss is bigger than the missing post.

The Failure Pattern

We saw the same pattern repeat:

• A queue existed in theory, but not always as an actionable file.
• One distribution channel looked available, then failed on auth or policy.
• The team assumed "we will publish later" instead of forcing a fallback path now.

In other words, the bottleneck was not content quality. It was operational ambiguity.

Rule 1: Minimum 2 Publish Actions Per Day

The first rule is simple:

Every day must contain at least 2 publish actions, not 1 intention.

Why two?

• One action can fail because of auth, moderation, or platform drift.
• Two actions force channel redundancy.
• A daily floor is easier to audit than a vague weekly goal.

This changed the conversation from "did we have a good idea?" to "which two channels did we actually attempt today?"

Rule 2: Empty Queue Means Create New Content Immediately

The second rule fixed the more dangerous case.

If the publish queue has no usable candidate, the system must not wait for permission. It must create one new piece immediately.

That content does not need to be perfect. It only needs to be real, current, and distributable.

Good fallback topics this week:

• PM weekly retro
• KIS or market infra notes
• MCP tool discovery
• blocked-channel lessons that turned into policy

This sounds obvious, but operationally it matters because an empty queue is now treated as a trigger, not a pause.

What Actually Improved

Three things got clearer:

1. Channel health is now part of product ops.
2. Publish count is an execution metric, not a marketing vanity metric.
3. Fallback writing is cheaper than waiting for the "best" post.

That last point matters most. Waiting preserves comfort. Publishing preserves momentum.

The PM Takeaway

A surprising amount of growth work is not creative at all. It is just turning soft intentions into hard rules:

• if queue missing -> inspect and recover
• if channel blocked -> reroute
• if no candidate -> write now

Once those rules exist, publishing stops depending on mood, memory, or optimism.

And that is the real win from this week: distribution became a system, not a hope.