Four tools. One real dataset.
Everything below runs on the same index: 1,376,116 Reddit posts and comments across 50 subreddits, including 3,293 high-confidence exact-SKU matches. Product rankings are deduplicated by user. No quiz walls, no email gates, no sponsored slots.
Quality gate: 28,980 ambiguous common-phrase matches are withheld from product evidence.
Reddit Finder
- Evidence source shown on every match; exact-match depth where available
- 6 curated intent guides for the big questions
- Exact-match quote where verified, plus a visible why-it-matched trail
Dupe Finder
- Savings math and evidence strength on every pair
- Co-discussion counts are labeled as context, not proof
- Direct comparisons and alternative-only context are labeled honestly
Search
- One quality-gated index shared with Finder and Catalog
- Evidence source labeled; exact-match quotes only where verified
- Load more, honest unavailable toggle, and ⌘K
Catalog
- Search plus category, type, concern, skin, price and sample facets
- Evidence source labeled; exact-match quotes only where verified
- 60-at-a-time rendering keeps the full catalog responsive
| You're thinking… | Use |
|---|---|
| "What should I buy for my skin?" | Reddit Finder |
| "This is great but $185 hurts." | Dupe Finder |
| "What does Reddit say about this specific product?" | Search |
| "Show me everything in one place." | Catalog |
How we count
Public database counts and quotes require an exact product phrase or at least two close, distinctive product tokens beside the brand. Broader brand and one-token matches stay in the discovery layer, but do not power SKU evidence. We then deduplicate by username and weight products by distinct redditors rather than raw comment volume. The full method is on the methodology page.
Products without a current exact match remain searchable with a clearly labeled catalog-aggregate fallback.