Start Building

Lesson 29: Two-Week Post-Patch Confidence Tracker for Yellow Mitigation Closure and Live Stability

Lesson 28 gave you a patch-readiness briefing packet with go, yellow, or red sign-off.

This lesson answers the next question: how do you prove yellow was temporary instead of discovering two weeks later that you shipped a slow-motion incident?

You will build a two-week post-patch confidence tracker that is small enough for a two-person team and strict enough for skeptical stakeholders.

Alien Character Designs illustration for post-patch confidence tracker lesson

What You Will Build

By the end of this lesson, you will have:

  1. A 14-day tracker grid with daily lanes for engineering, support, and telemetry
  2. A yellow closure rule that defines what counts as done for each mitigation
  3. A live stability gate that compares week one versus week two variance
  4. A promotion retrospective hook that feeds Lesson 27 retro format without duplicating it

Step 1 - Copy mitigation rows from the Lesson 28 packet

Open your most recent patch-readiness packet from Lesson 28: Patch-Readiness Briefing Packet for Go Yellow Red Sign-Off.

Copy only the rows that shipped as yellow with an owner and due date.

If nothing was yellow, you still run the tracker at reduced intensity: one daily stability row and one dialogue health row for AI-heavy patches.

Step 2 - Define closure criteria before day one

For each yellow mitigation, write one observable closure line:

  • Engineering example: crash signature X below threshold for 7 consecutive days
  • Support example: refund spike topic cluster below baseline for 5 consecutive days
  • AI dialogue example: fallback rate below agreed percent with zero P0 safety flags

Pro tip: If you cannot measure it, it is not a mitigation. It is a hope.

Common mistake: Treating "we deployed a fix" as closure without measuring player-facing symptoms.

Step 3 - Build the 14-day tracker table

Use one shared sheet with columns:

Day Build ID Eng signal Support signal Telemetry signal Yellow items open Notes

Keep notes to one sentence so the table stays readable on mobile.

Step 4 - Week one versus week two interpretation

Week one validates that the patch did not introduce immediate regressions.

Week two validates that mitigations actually converged instead of masking drift.

If week one is green but week two degrades, treat that as a telemetry or caching investigation, not bad luck. Cross-check the drift guidance from Lesson 26: Analytics Confidence Review and Live-Ops Handoff.

Step 5 - Close the loop with launch operations patterns

When your RPG ships alongside storefront operations, align language with Lesson 22: Post-Launch Stabilization Sprint Board so support and engineering use the same severity vocabulary.

If you need a weekly commercial read during the same window, reuse the scorecard rhythm from Lesson 24: Monthly Launch-Ops Scorecard and Decision Rhythm at half depth for those fourteen days only.

Mini Exercise

Create post_patch_tracker_patchID.csv with fourteen rows pre-filled dates and empty signal cells.

Add three yellow closure lines copied from your Lesson 28 packet.

Troubleshooting

Yellow items close early but players still complain

You measured the wrong proxy. Re-open the mitigation row and tie closure to a player-visible metric.

Telemetry looks noisy after a CDN or config change

Freeze unrelated experiments during the tracker window so you are not chasing multiple moving variables.

Support volume is low but quality is bad

Switch one column to sentiment or refund reason tags instead of raw volume.

FAQ

Do we still run this if we signed green

Yes, but shorten to seven days and keep only stability plus AI dialogue health rows.

What if we hotfix during the two weeks

Reset the tracker from the hotfix build ID and carry forward any unresolved yellow rows with explicit carry-over notes.

Who owns the daily update

One rotating owner is enough if everyone agrees the update time and the evidence links are pre-defined.

Lesson Recap

You now have a two-week post-patch confidence tracker that:

  • proves yellow mitigations actually close
  • compares week one and week two stability honestly
  • keeps evidence tight enough for small teams
  • connects AI RPG operations to broader launch-ops habits

Next Lesson Teaser

Continue with Lesson 30: Multi-Patch Rolling Stability Dashboard for Release-Train Aggregation to roll closed tracker columns into a small release-train view without duplicating your Lesson 28 briefing packet.

Related Learning

Bookmark this lesson and start the tracker the same day you promote a yellow-touched build.