Why a selection policy exists

A price index is only as honest as its inputs. The most predictable way to distort one is to get the wrong inputs added: a storefront with prices that look low but cannot actually be transacted, or inventory that duplicates stones the index already tracks, would drag published medians away from what a real shopper can actually pay. This page states, in advance and in public, what it takes for a retailer to be considered, so that every decision about the panel is a policy decision rather than a negotiation.

Criteria a retailer must meet before being considered

• Operating history. At least 12 months selling loose lab-grown diamonds to consumers under the current brand, verifiable through public records, archived site history, and independent reviews.
• Verifiable business identity. A registered legal entity locatable in a public registry, a physical business address that is not a mail-forwarding service, and a meaningful independent footprint such as customer reviews, press coverage, or industry presence.
• Independent inventory. Many lab-grown listings originate in shared wholesale feeds, so the same certified stone can appear at several storefronts at once. We measure inventory overlap against the existing panel by certificate number. A storefront whose book substantially duplicates an existing panel retailer's certificates adds double-counting, not information, and does not receive a scored row.
• Public, stable, per-stone prices. Prices visible to any shopper without a login, trade account, or quote request, with per-stone certificate numbers from an independent grading lab that can be verified at the lab itself.
• Fulfillment credibility. Evidence that the retailer actually transacts at its listed prices, such as a sustained independent review history. Listed prices that cannot be transacted are not market data.
• Data quality over time. Inventory and pricing that survive our sanity checks, with no implausible prices and no wild unexplained swings, across at least three consecutive monthly collection cycles that we run ourselves before any decision is made.

How a qualifying retailer would actually join

• We observe the retailer silently, with our own collection method, for a minimum of three monthly cycles before any decision. Nothing a retailer sends us substitutes for our own collection.
• If added, a retailer debuts as a disclosed new addition from a stated month forward, never retroactively. A new retailer may first appear as a clearly labeled reference row before its figures ever enter any published median.
• Every addition to or removal from the panel is disclosed in the methodology notes of the next monthly edition.

A note for retailers

We periodically re-evaluate the retail landscape as part of the research process. That process requires no action from retailers: there is no application, and inclusion requests are not acted on.