Steam Store FAQ Lines and Demo Scope Parity - Next Fest 2026 Checklist for Micro-Studios

Players open your FAQ when something already feels wrong. They compare your "Does the demo include multiplayer?" answer to hour one in the zip. They screenshot "Full game at launch" beside a fest build that ends at the tutorial gate. Partners treat FAQ rows like contract language under intake compression—especially when store-demo mismatch recovery timelines start with FAQ paste, not gameplay bugs.

This Process & Workflow checklist audits every FAQ question and answer against Next Fest demo hour-one proof. It pairs with the FAQ LLM draft pipeline (how drafts get written) and the seven-day metadata sprint (when FAQ fits Day 7 integration). Anchor expectations to Steamworks store page documentation. No invented studios or metrics.

Time to read: ~34 minutes. Time to run first FAQ audit: ~2.5 hours with cold install, export, and proof table.

Why this matters now (May–October 2026)

1. FAQ is the refund script — Support macros and player anger start with FAQ rows, not About paragraphs.
2. LLM drafts amplify FAQ count — Teams add twelve new Q&A rows in May; few get hour-one proof.
3. Fest traffic is FAQ-heavy — Install failures route to "system requirements" and "demo content" answers you wrote for full game.
4. Mock audit dimension 7 — Tabletop seven dimensions scores FAQ against demo and disclosure.
5. May is the rewrite month — October is typo-only unless scope changes; FAQ drift in June still breaks Wednesday diffs.

Direct answer: Export live FAQ to git, build a faq-line-proof.md table (one row per Q&A pair), mark pass/fail against cold-install hour one on current build_id, rewrite or delete failures, and diff weekly like other G-surfaces.

Who this is for

Prerequisites

Optional: run wishlist truth audit first if you never audited store copy.

FAQ vs other store surfaces (do not duplicate work)

If FAQ contradicts any green surface, fix FAQ first unless the other surface is wrong—then reopen that checklist.

FAQ line proof table (core artifact)

Create release-evidence/03-store/faq-line-proof.md:

Pass rule: Engineer reproduces the answer claim without dev cheats, mods, or debug menus on the fest zip tied to current build_id.

Fail rule: Rewrite answer, split into DEMO vs FULL GAME sections, or delete the row. Never leave a fail row live into October.

High-risk FAQ categories (audit every row)

Scope and content

Questions about what is in the demo, how long, chapters, endgame, DLC, season pass. These cause the most refunds when wrong.

Multiplayer and social

Co-op, PvP, cross-play, dedicated servers, friend invite. Pair with tag coherence—if tags removed co-op, FAQ must not keep it.

Platforms and input

Deck, Mac, Linux, controller, Steam Input. Each answer needs a binary or config proof row. Use Deck tools listicle when claiming handheld play.

AI and generative content

Any FAQ mentioning AI art, voice, procedural narrative, or ML-assisted balance must match storefront disclosure and prompt registry if runtime AI exists.

Legal and commercial

Refunds, early access date, price, regional availability. Financial FAQ may pass without hour-one gameplay proof only if the answer does not imply demo content that is not shipped.

Technical support

Crashes, antivirus, cloud saves, minimum specs. Answers must match spec rows and crash-log hygiene artifacts.

Question templates that usually fail (delete or rewrite)

1. "What features are planned?" — Unless you label FULL GAME and separate DEMO section.
2. "Is this the full experience?" — Requires explicit demo boundary in the answer.
3. "Can I play with friends?" — Default fail for solo demos; rewrite to "Demo does not include multiplayer."
4. "Will progress carry over?" — State save compatibility honestly or remove.
5. "Unlimited procedural content" — Ban unless engineer proves unbounded generation in hour one.
6. "All languages at launch" — Match localization QA scope.
7. "Same content as trailer" — Requires trailer frame table reference.
8. "Roadmap Q3 2026" — Move to news post; FAQ is not a roadmap billboard.

Answer writing rules (demo-safe)

Pairing with the FAQ LLM pipeline (Gate E)

If you use LLM drafts:

Never skip Gate E because Gate D "looked fine." Models hide scope in polite filler words.

Wednesday FAQ export ritual (fifteen minutes)

Add to Wednesday metadata diff:

1. Export live-faq-YYYY-MM-DD.txt
2. diff live-faq.txt store-copy/faq-approved.md
3. If diff non-empty, open proof table before promotion
4. Log result in wednesday-diff-log.md row FAQ

FAQ changes without diff are how tag drift happens in text form.

Integration with metadata sprint and freeze

BUILD_RECEIPT fields to extend

Add to your receipt JSON:

"faq_surfaces_locked": false,
"faq_row_count_live": 0,
"faq_proof_table_path": "release-evidence/03-store/faq-line-proof.md",
"faq_last_export_sha256": ""

Set faq_surfaces_locked true only when proof table is all green and producer signed.

Evidence folder layout

release-evidence/03-store/faq/
  README.md
  faq-line-proof.md
  live-faq-YYYY-MM-DD.txt
  faq-promotion-log.md
  engineer-cold-install-YYYY-MM-DD.log
  producer-signoff.txt

Zip this folder into partner upload packets under store marketing truth.

Red-team questions (tabletop)

Run these aloud before October:

1. "I read FAQ #4—where do I click co-op?"
2. "FAQ says 10 hours—why did I finish in 2?"
3. "FAQ promises Steam Deck Verified—why is glyph wrong?"
4. "FAQ mentions mods—where is the mod menu?"
5. "FAQ says AI voices—where is disclosure on the store page?"

Any stumble → fail row → rewrite before ads restart per fest marketing cap.

Triangle diff (FAQ leg)

On integration day, cross-check:

Same triangle as metadata sprint Day 7—FAQ is the third leg partners quote in emails.

Publisher diligence one-pager

Append to diligence README:

FAQ parity audit <date> — faq-line-proof.md green for build_id <id>; live export SHA256 logged; LLM Gate D sign-off attached if AI-assisted.

Supports publisher diligence packets without a live call.

When to re-audit FAQ only

Common mistakes (twelve)

1. Treating FAQ as support blog—not binding copy.
2. Copy-pasting About bullets into answers without proof rows.
3. Adding FAQ rows to reduce tickets—scope creep.
4. Skipping engineer sign-off because "it's just text."
5. Leaving "full game" answers in demo FAQ section.
6. Using LLM without frozen source bullets.
7. No export after Steam paste.
8. Contradicting spec honesty in crash FAQ.
9. Promising event rewards not in build.
10. Ignoring two-storefront rule for itch HTML5 FAQ clone.
11. October typo edits that secretly add modes.
12. No validate-packet FAQ dimension before partner ZIP.

Worked example (anonymous)

Starting state: fourteen FAQ rows; three fail co-op and hours claims.
Tuesday: Cold install; proof table built; rows 2, 7, 11 fail.
Wednesday: Rewrite row 2 to "DEMO: solo only"; delete row 7 roadmap; narrow row 11 to "90–120 minutes main quest."
Thursday: Export + empty git diff; faq_surfaces_locked true.
Friday: validate-packet exit 0.

Engineer time ~3 hours; writer ~4 hours; producer ~1 hour.

Regional and DLC FAQ annex

If you sell in multiple regions or mention DLC:

• Regional pricing FAQ must not promise features by currency tier.
• DLC anchor FAQ must label FULL GAME purchases separately from DEMO.

Add thirty minutes to proof table—do not merge DLC answers into demo content rows.

Post-fest maintenance

1. Wednesday FAQ diff forever.
2. Friday Block 5 FAQ row each week per operating review.
3. Demo patch notes after scope changes—link from changed FAQ rows.
4. Evidence cycles batch non-critical wording weekly per evidence cycles opinion.

Glossary

First audit sprint (ninety minutes, solo-friendly)

Use this sequence the first time you touch FAQ this season:

If more than three rows fail, stop paste work and schedule a half-day rewrite—do not ship a partial table.

Sample proof rows (copy structure)

Row — Saves and cloud:

Row — Performance:

Row — Content parity:

Ordering FAQ rows for skim logic

Steam displays FAQ order you set in Steamworks. Recommended stack for fest demos:

1. What is included in the demo? (boundary first)
2. How long is the demo? (hours or missions)
3. Multiplayer / co-op (explicit no if solo)
4. Platforms and Deck (binary truth)
5. Known issues / crashes (link logs path)
6. AI / disclosure (if applicable)
7. Full game / Early Access (FULL GAME labeled)
8. Price / refunds (commercial, no scope creep)
9. Technical requirements (pointer to spec rows)
10. How to report bugs (support email, not scope)

Putting boundary rows first reduces refunds from players who never scroll to row nine.

validate-packet FAQ dimension (Day 7)

Extend your validate-packet script with optional checks:

• faq-approved.md exists and is non-empty
• faq-line-proof.md has zero Fail rows (case-sensitive)
• Live export SHA256 matches receipt field
• Row count in export ≤ approved row count + 0 (no surprise Steam-only rows)

Exit non-zero blocks partner ZIP assembly—same discipline as hash and depot guards.

Ads, UTM, and FAQ coupling

If UTM experiments send traffic to FAQ anchors, freeze FAQ edits during the experiment week. Changing answers mid-experiment confounds retention data and can violate ad network truth expectations when landing copy references FAQ quotes.

Asia–EU handoff note

For teams using Asia–EU evidence handoff, leave faq-line-proof.md in the shared folder with open fails highlighted—do not let overnight writers paste Steam FAQ from memory. Morning EU review starts with export diff, not inbox tone.

Comparison to reactive recovery

FAQ lies are the fastest path to the middle column because players quote answers in tickets partners forward verbatim.

Team RACI (lightweight)

R = responsible, A = approves, C = consulted, I = informed.

Metadata tools cross-link

When exports fail, use 14 free metadata export tools before screenshotting Steamworks by hand—hand copies introduce typos that fail diff gates without failing truth.

Schedule the first full FAQ proof table the same week you green-light BUILD_RECEIPT—receipt culture without FAQ rows is half a governance story. Treat every new FAQ row added before October as a scope claim that needs the same cold-install discipline as a new gameplay feature flag.

Key takeaways

• Every FAQ Q&A pair needs a proof table row tied to build_id.
• High-risk categories: scope, multiplayer, platforms, AI, legal, tech support.
• Use DEMO / FULL GAME labels in answers—players and partners read literally.
• Pair with FAQ LLM pipeline Gate E before paste.
• Export live FAQ weekly; empty git diff before promotion.
• Triangle diff FAQ against short description, tags, and trailer.
• Evidence lives under release-evidence/03-store/faq/.
• Red-team five questions before October ads.
• Cheaper than 48-hour mismatch recovery.
• Completes textual metadata cluster beside About and short description.

FAQ

How many FAQ rows should a fest demo page have?
Enough to kill predictable refunds—often 8–14. More rows mean more proof work.

Can we use the LLM pipeline without this checklist?
No—Gate E is parity scan; this checklist is the pass/fail definition.

What if FAQ matches itch but not Steam?
Run two-storefront rule—separate proof tables.

Does Steam enforce FAQ accuracy?
Players and partners enforce it via refunds and yellow flags.

Where does FAQ sit in mock audit?
Dimension 7 tabletop—bring faq-line-proof.md.

October edit policy?
Typos only unless build_id changes—then re-run proof table.

Conclusion

FAQ lines are where polite marketing becomes enforceable promises. A fourteen-row FAQ is fourteen chances to recreate store-demo mismatch without touching a single line of gameplay code.

Export live copy, build the proof table, and let engineers fail rows before players do. When October traffic arrives, your FAQ will read like support documentation that already told the truth—because you proved it on floor hardware in May.

Run the table this week. Delete one overpromising answer. Everything else in your metadata cluster gets easier when FAQ stops lying first.