Lesson 14 locked balance with humans in the loop. This lesson is the handoff—a binary reviewers can run, a trailer that shows your best beats without spoilers, and a case study page that survives when the project zip goes stale.

Lesson objective

By the end of this lesson you will have:

1. A packaged Shipping build (or Shipping game target with known caveats) with maps, plugins, and staging rules documented next to the exe.
2. A 45 to 60 second trailer plan executed in Sequencer (or replay markers) with fixed cameras, no player wobble, and mix that matches Lesson 11 readability.
3. A one-page case study (PDF or markdown) that names Lesson 1 pillars, lists two metrics from playtests, and shows one honest cut feature.

Step 1: Packaging preset that matches your promise

1. Duplicate your packaging profile: StealthSlice_Shipping vs the Development profile you used in Lesson 13.
2. Full Rebuild path clean once before the golden package—stale cooked content is the silent killer of demo night.
3. Map list: only include what ships; strip greybox diagnostic maps unless they are the slice.
4. Plugins: disable editor-only tools; confirm Online, EOS, or Steam stubs match your actual slice (offline is fine—lies in packaging settings are not).

Pro tip: Zip with README_build.txt inside next to the exe—controls, known bugs, build ID, and GPU tier you validated.

If packaging fails on missing module or plugin, use our UE 5.5 packaging troubleshooting before you chase Blueprint errors.

Step 2: Staging and size sanity

1. Project Settings → Packaging → List of maps to include explicitly for deterministic builds.
2. Internationalization: cook only languages you ship; drop unused loc assets.
3. Run cook on command line once if Editor packaging hides warnings—capture the log in Saved/Logs.
4. Compare folder size to Lesson 13 budget; if Shipping balloons, diff against Development staging to find accidental 4K refs.

Step 3: Trailer rails (deterministic, not gameplay luck)

1. Pick three beats only: setup (fantasy), tension (detection near-miss), payoff (objective clear or narrow escape).
2. Place Cine Camera Actors on splines or static keyframes in Level Sequence; match Lesson 12 exposure locks so HDR does not pump between cuts.
3. Record at fixed frame rate (30 or 60) with motion blur policy decided up front—do not change mid edit.
4. Audio stem export or bus bounce if you edit in DaVinci / Resolve; keep UI muted unless detection read is the hero beat from Lesson 10.

Pro tip: One hero shot from AI search state reads better than five fast cuts of crouch walking.

Step 4: Case study one-pager (portfolio-safe)

Mirror Lesson 1 structure:

1. Title + one sentence hook (who you are, what the slice proves).
2. Pillars (three bullets) copied or refined from Lesson 1—no new lore here.
3. Systems shipped (perception, objectives, gadgets) with engine labels a hiring manager recognizes.
4. Metrics (median time-to-detect, win rate, quit points) from Lesson 14—if you have no numbers, say what you will measure next and why.
5. What you cut (scope discipline) and what you would fund next (honest roadmap).

Step 5: Ship checklist (copy into your repo)

• Build ID + git hash or changelist
• Target GPU / FPS from Lesson 13
• Map name to launch first
• Controls layout image or text
• Credits for audio / fonts / marketplace assets
• Contact or Itch / Steam placeholder link if public

Recap

• Shipping settings are part of design—they encode what you claim works.
• Trailers sell timing; rails beat lucky runs.
• Case studies translate engine work into hire language without overselling.

Further reading

• Publishing: Packaging and Distribution in Unreal — platform nuance beyond this slice.
• Performance Optimization and Profiling in Unreal — last look before you encode trailer shots at the wrong scalability tier.

FAQ

Shipping vs Development for reviewers?
Shipping for publishers and portfolio links; Development only if you need console diagnostics and you label it clearly.

Can I use Take Recorder instead of Sequencer?
Yes if you trim input noise and stabilize cameras after; Sequencer usually wins for repeatable exports.

How long should the case study be?
One page or one scroll—recruiters skim; depth lives in appendix links.

What if my slice still has placeholder art?
Say so in the case study and trailer credits—honesty beats a polished lie.

Next: You have finished this course track. Polish your build, upload the trailer, and iterate the case study after every playtest round. For another end-to-end ship path, continue with Build a Complete Indie Game in Unity 2026 or Build a Godot 4 Action Adventure from Scratch when you want a different engine rhythm.