Premium Comparison Pages

Every /compare/[slug] URL (e.g. /compare/hubspot-vs-salesforce, /compare/outreach-vs-smartlead, /compare/clay-vs-zoominfo) renders from a PremiumComparisonCopy entry in lib/compare/premium-comparisons-registry.ts. Each entry has ~170 lines of structured editorial: a five-dimension quickVerdict (SMB pick, enterprise pick, data, ease-of-use, hidden cost), a six-row comparison table, deep analyses of each tool (what it does / where it shines / where it breaks / typical stack), whatTeamsGetWrong bullets, cost-reality paragraphs, a StackSwap-angle section positioning StackScan as the diagnostic, final verdict paragraphs, best-alternatives internal links, when-both-rare guidance, and four FAQ Q&As. The page emits Article + FAQPage JSON-LD via lib/seo-jsonld.ts and a canonical URL + OpenGraph metadata.

The /knowledge/comparisons hub lists 13 curated head-to-heads across CRM, sales engagement, enrichment, marketing automation, and conversation intelligence. Each card links to the matching premium page. New comparisons are a single registry entry — no custom layout, no new route, no bespoke metadata plumbing. The canonical tool registry (lib/tools-data/records.ts) supplies score pills and cross-links, so adding a comparison for a new tool pair requires (1) the two tool records, (2) a comparison slug, (3) the premium copy block. Everything downstream (sitemap, JSON-LD, breadcrumbs, related comparisons) derives automatically.

Operators evaluating "Outreach vs Salesloft" or "Apollo vs ZoomInfo" increasingly ask LLMs first, Google second. Structured FAQPage + Article JSON-LD makes StackSwap's take extractable in AI Overview surfaces; the quickVerdict pattern gives LLMs clean Q&A snippets to cite. The editorial voice is operator-first — tradeoff-honest, pricing-realistic, and explicit about when the comparison is the wrong question (consolidate instead). That's the kind of content both humans and language models pick up as authoritative.