Methodology

How We Review Jewelry

The criteria, scoring, and refresh cycle behind every recommendation we publish. The standards apply equally whether a link is affiliated or not.

Updated

What "reviewed" actually means here

JewelryHubFinder is a discovery and buying-guide platform — we don't physically test every individual SKU on every list (no jewelry editor does honestly). What we review is the supplier, the spec sheet, the certification, the return policy, and the public reception of the design over at least 90 days of sales data.

When a guide includes a recommendation, it means the supplier, the spec, and the design have all passed our criteria. It does not mean we've worn the specific piece for months. We're explicit about this because most "reviewed" jewelry online uses the word in a way that's closer to "promoted."

The four scoring criteria

1. Material quality and disclosure

The seller must publish the karat, fineness, alloy composition, and (where applicable) the grading certificate for any stone over 0.5 carat. Vague terms — "solid gold," "precious metal," "genuine diamond" without grading — drop a piece off the list. For our take on what to verify, see our buying-guides pillar.

2. Real-world durability

Will the piece hold up to daily wear? Settings (prong, bezel, channel), metal karat, clasp type, and joint construction all factor in. Pieces with known structural weaknesses — exposed pavé on the bottom of a ring, thin bezel on a daily-wear pendant — get noted in the guide or excluded.

3. Value at the budget

We compare price against two benchmarks: similar-spec pieces from other retailers, and the loose stone-plus-setting cost if commissioned independently. A piece priced 30%+ above benchmark needs a clear reason to stay on the list (brand heritage, design exclusivity, exceptional workmanship).

4. Seller policy

Return window (30 days minimum), free resize policy (at least one within 60 days for rings), shipping insurance, and resolution history on common issues like stones loosening or chains breaking. Sellers with consistent reader complaints get pulled regardless of how good the spec sheet looks.

How often we update

Every published list is re-checked at least quarterly. We verify:

  • Current retail price and any price changes since the last update
  • Whether the recommended SKU is still in stock and shipping
  • Whether the return policy or warranty has changed
  • Whether the brand has had material reader complaints during the period
  • Whether better-priced or better-quality alternatives have launched

Picks that have slipped on any of these get replaced. The visible "Updated" date on each guide reflects the last full re-check, not just a minor copy edit.

What we won't do

  • Accept payment for inclusion or ranking position
  • Publish AI-generated "reviewer quotes" or synthetic user testimonials
  • Recommend a piece we couldn't verify the spec on
  • Rank pieces by affiliate commission rate
  • Keep a stale pick on a list because removing it would tank affiliate revenue
  • Publish thin doorway pages designed only to rank — every guide must add real reader value

Related reading