Product Case Study

Duet Night Abyss — GachaBuild3

A bilingual game database and guide built to close the gap between scattered player discussions, stale patch data, and the need for fast, answer-first character decisions.

Duet Night Abyss project preview
Duet Night Abyss tier list screenshot
120+
Organic keywords
~38%
Tier list CTR
~99.8%
Uptime
EN / VI
Bilingual delivery
Strategic Vision

The product closed a gap that every competing wiki ignored.

When Duet Night Abyss launched in 2024/2025, the global and Vietnamese player communities suffered from fragmented tier lists, absent bilingual support, and stale patch data. GachaBuild3 was built to close this gap within 90 days — establishing first-mover authority in both language markets.

“The definitive English + Vietnamese DNA encyclopedia — built by the community, for the community.”
Vision

Become the #1 EN/VI knowledge base

Deliver data-verified character rankings, build guides, and weapon analytics in both English and Vietnamese.

Value proposition

Authoritative answers in under 3 seconds

The only bilingual, real-time-updated, mode-ranked tier list and character database. Every page optimized for the exact question a player is asking.

Pain Points

Players needed confidence, speed, and trustworthy localized answers.

The biggest product opportunity was not another wiki. It was reducing uncertainty at the exact moments players had to make roster and resource decisions.

Critical

Decision paralysis

New players pulled a 5-star character and had no fast way to know if it was actually meta — scattered Reddit threads and Discord debates offered no consensus.

Critical

Language barrier

60%+ of DNA's SEA player base is Vietnamese; English wikis left them behind. Machine-translated community posts created frequent inaccuracies.

High

Stale patch data

Community spreadsheets and hobby wikis lagged behind balance changes by days or weeks. Players made roster decisions on outdated truth.

High

Fragmented builds

Weapon pairings, team comps, and skill priorities were scattered across Discord and YouTube instead of searchable, structured data.

Product Pillars

Three operating pillars turned a fan problem into a production system.

Strategy

Product

A mode-ranked, answer-first experience. Homepage opened on highest-intent answer: S/A/B/C tier bands scannable in under two seconds.

Content

Content Ops

Sanity schemas with EN + VI inline fields. Webhook-driven ISR revalidation kept rankings fresh in under five seconds.

Engineering

Engineering

Next.js App Router, static search, Docker Compose, Nginx, GA4, and SEO hardening created a production system that ran itself with 99.8% uptime.

Pillars Decision
“The most consequential decision was pairing schema-level bilingual content with ISR revalidation. That made freshness and accessibility native capabilities of the product — not cleanup work after launch.”

— Duet Night Abyss Product Pillars

System Architecture

A lean stack optimized for freshness, SEO, and low-maintenance operations.

Next.js 15 · App Router

SSR, SSG, and ISR with React Server Components — client JS only where interactivity mattered.

Sanity CMS v4

Structured content with real-time editing, GROQ queries, and automatic webhook revalidation pipelines.

Docker + Nginx

3-service production stack with SSL termination, path-based routing, and unless-stopped restart policy.

React 19

Concurrent features, improved SSR hydration, and Context-based EN/VI language state management.

TailwindCSS 3.4

Utility-first styling with custom professional-card patterns and a Geist type system for zero layout shift.

GA4 + SEO

Bilingual hreflang, JSON-LD structured data, per-page OpenGraph, and 6 custom event trackers.

Results & KPIs

Early traction proved the value of bilingual, answer-first product design.

~99.8%
Launch Uptime

3-service Docker stack with unless-stopped restart policy

~1.8s
Largest Contentful Paint

Fast enough for mobile-first organic traffic

120+
Organic Keywords

Strong early bilingual SEO footprint in 30 days

~38%
Tier List CTR

High clickthrough from scannable rankings into details

~15–20%
VI Adoption

Meaningful bilingual usage from VN audiences

< 5s
ISR Latency

Fresh content shortly after CMS edits

Legacy

Decommissioning is part of product ownership.

This project closed with the same discipline it used to launch.

Legacy

Content preserved

Export Sanity content, images, and final structured records via CSV and JSON before shutdown.

Legacy

Analytics retained

Preserve GA4 data, uptime notes, and operational context for future audits or case-study verification.

Legacy

Domain handled cleanly

Point the project toward a static archive if discoverability still matters after decommissioning.

Legacy

Secrets revoked

Invalidate API tokens, webhook secrets, and deployment credentials after the final takedown.

Back to Portfolio

Explore the rest of the work.

Return to the homepage for the full portfolio, PM journey, and supporting work.

Back to Portfolio