All Projects
In Development2026

RoastRank

The same coffee scores differently as espresso vs pour-over — RoastRank captures that

A coffee discovery platform whose core differentiator is brew-method-specific rankings, built on a statistical scoring model that keeps small review counts from skewing results.

7
Brew methods ranked independently
1,000
Coffees across 300 roasters
12
Ranking pages built for search traffic
32
Flavor tags in a two-tier hierarchy

Best Coffees for Espresso

Bayesian weighted · minimum 5 reviews per brew method

EspressoPour OverAeroPressCold Brew
1

Ethiopia Yirgacheffe

Luna Roast Works

BerryFloral

9.2

214 reviews

2

Colombia Huila

Harbor & Vine Coffee

CaramelStone Fruit

9.0

178 reviews

3

Kenya Nyeri AA

Northbound Roasters

CitrusSweet

8.8

96 reviews

Same coffees, different order under Pour Over — that's the point.

Illustrative mockup — representative of the product interface, with placeholder data.

Overview

Coffee review sites collapse everything into one number, but a coffee that shines as espresso can fall flat as pour-over. RoastRank segments every rating by brew method — espresso, pour over, drip, French press, AeroPress, moka pot, cold brew — and ranks coffees separately for each.

Reviewers leave a quick 1–5 star rating with their brew method, or go deeper with a detailed rating across four tasting dimensions — brightness, sweetness, fullness, intensity. Rankings use a weighted average so a coffee with three perfect reviews can't leapfrog one with two hundred strong ones.

Tech Stack

Next.js 15 (App Router)TypeScriptPrisma + Neon PostgreSQLTailwind CSS + shadcn/uiNextAuthMDX

The Challenge

Ranking integrity with sparse data: splitting reviews seven ways by brew method means each segment has fewer reviews to work with, so the scoring model has to stay trustworthy even with a small sample.

The Approach

Every rating feeds a documented scoring formula, followed by a statistical weighting step (Bayesian averaging) with minimum-review thresholds — five reviews per brew method, ten overall — so a handful of glowing reviews can't game the rankings, and new coffees still get discovered.

The platform is built for organic growth: twelve ranking pages targeting search traffic refresh every fifteen minutes, an editorial Discover hub hosts brew guides and curated lists, and affiliate links with click tracking provide the revenue layer. Community contributions are reputation-gated — submitting coffees, editing details, and approving edits unlock as contributors earn points.

How a Rating Becomes a Ranking

  1. 01ReviewStars + brew method, quick or detailed
  2. 02Composite ScoreFour tasting dimensions, weighted
  3. 03Fair WeightingA few rave reviews can't skew results
  4. 04Brew RankingsSeparate lists for all seven methods

Highlights

  • Brew-method-specific rankings — the product's defensible core
  • Quick star ratings or detailed four-dimension tastings — both feed the rankings
  • Reputation-gated community: submissions, edits, and approvals unlock as contributors earn points
  • 12 ranking pages built to win search traffic, refreshed every 15 minutes
  • Affiliate click tracking with referrer analytics
  • Editorial content hub: articles, brew guides, and curated lists

Want something like this built?

We design and build AI products, analytics platforms, and automation systems end to end. Tell us about your problem.