The next growth phase: ~250 statically-generated programmatic scam pages (by type, city, bank, UPI app, platform), AI-Overview/Discover/featured-snippet formatting, an authority + citation system tied to 1930/cybercrime.gov.in, an internal-linking engine, and India-first bilingual GEO — all on a lightweight, Vercel-hobby-friendly, near-zero-cost architecture. Includes highest-ROI opportunities, traffic-growth potential, and fastest-ranking keywords.
This phase turns ScamCheck + TrustSeal from single-page tools into a topical authority on scams in India, using programmatic SEO that is structurally optimized for Google's AI Overviews and Discover — at near-zero running cost.
The consumer apps are client-rendered Vite SPAs on GitHub Pages. Client-only pages are not reliably indexed, cited by AI Overviews, or eligible for Discover. The new pages are statically generated (SSG) in the Next.js engine — real HTML at build time, zero runtime cost — served at /scams/... (lab today; map a scam subdomain via NEXT_PUBLIC_SCAM_BASE_URL when ready).
Generated deterministically from a structured India-first data layer (lib/seo/facets.ts) — no AI cost, no hallucination:
directAnswer lead block (≤320 chars) → verdict snippet → quick bullets → FAQ-first (FAQPage JSON-LD) → structured summary table. This is the exact shape answer engines lift.
Honest, scannable headlines (no clickbait), truthful emotional hooks, freshness labels, max-image-preview:large, and a ≥1200px OG image recommendation per page (lib/seo/discover.ts).
Every page cites real sources — helpline 1930, cybercrime.gov.in, RBI Sachet, SEBI, Sanchar Saathi — plus a transparent trust score badge (lib/seo/authority.ts). Strong E-E-A-T → higher ranking + citation odds.
Each page links to related scams, trending (tier-1) scams, facet cross-links, and the live alerts hub (buildInternalLinks) — building a tight topical cluster.
Bilingual: English + हिन्दी direct-answer/verdict on every page; multilingual embeddings; India-first terms (₹, UPI, 1930). Default region India.
Plan for the consumer apps (the AdUnit already gates paid users): in-content placement after the verdict block (highest viewability), lazy-load below the fold, one anchor + one in-article unit max (avoid intrusive layouts → protects Core Web Vitals + RPM), auto-removed for paid users. Implementation is the next step.
SSG (no per-request compute), zero AI on the programmatic path, sitemap auto-includes all pages, Firestore untouched by these pages, Vercel-hobby compatible.
~250 pages × a conservative 15–60 clicks/page/month at maturity (3–6 months to rank long-tail) ≈ 4k–15k incremental organic clicks/month, compounding as topical authority builds and Discover/AI-Overview placements land. Tier-1 type pages can individually exceed this range. Estimates, not guarantees — depend on indexing, competition, and link signals.
/api/og rendering for the recommended 1200×630 images (Discover/social CTR).Programmatic pages add $0 AI cost (deterministic) and $0 Firestore (static). Build time grows linearly (~250 pages prerender in seconds). Fully Vercel-hobby compatible. The only ongoing AI cost remains the autonomous alert pipeline, already capped and cached (see low-cost doc).