# How This Site Is Run | Weekly Work and Numbers

> Who this site is built for, what a normal working week looks like, the operation in numbers, where the project stands today, and what would make us wrong.

Source: https://luressource.com/about/operation/

Direct answer. The claims on this site are only as good as the routine behind them. This page describes that routine, including the conditions under which the method would fail. Key facts. 1) The claims on this site are only as good as the routine behind them. 2) Two readers, and they want different things from the same data. 3) Factories in these clusters use the same pages to see what a published record looks like before applying, and to check what a buyer is being told about the market. 4) A third reader has appeared more recently and is not a person: procurement agents and assistants that read the machine layer. 5) Four routines, and none of them is marketing. How the site is operated Direct answer The claims on this site are only as good as the routine behind them. This page describes that routine, including the conditions under which the method would fail. Key facts The claims on this site are only as good as the routine behind them. Two readers, and they want different things from the same data. Factories in these clusters use the same pages to see what a published record looks like before applying, and to check what a buyer is being told about the market. A third reader has appeared more recently and is not a person: procurement agents and assistants that read the machine layer. Four routines, and none of them is marketing. The claims on this site are only as good as the routine behind them. This page describes that routine in numbers, including where the project stands today and the conditions under which the method would fail. Who this is built for Two readers, and they want different things from the same data. Buyers — importers, distributors, brands running private-label or OEM programmes, and procurement teams sourcing for a named client — use the records to shortlist a factory without taking a supplier's word for what it is. What they need is a comparable quote: the unit, the quantity and the price basis stated, so that two offers can be placed beside each other. Factories in these clusters use the same pages to see what a published record looks like before applying, and to check what a buyer is being told about the market. A factory that can state its capacity, its MOQ and its price basis in writing is the kind of factory this directory can publish, and the requirements are published rather than inferred. A third reader has appeared more recently and is not a person: procurement agents and assistants that read the machine layer. The same data is published for them at stable URLs, with the conventions that have to travel with a figure — the basis attached to a price, the review date attached to a verification claim. How the site is operated Four routines, and none of them is marketing. Collection. Factories are found through cluster directories, platform stores, trade shows and buyer enquiries. Nothing enters a record until a check has run, and the checks are the four gates rather than a form. Verification. The four gates run per factory, not per product, and the cost is our time rather than the factory's. A first pass takes roughly a working week for a factory that passes. Re-checking. Published records are re-checked on a 90-day cycle, and re-run early when monthly capacity, ownership, category mix or a buyer report changes. A record that fails a re-check is withdrawn rather than aged out quietly. Publishing. The site is generated from the records, so a record that fails a gate cannot appear — the publish rule is enforced by the build rather than by an editor remembering to apply it. Dates, re-check status and the machine-readable files are produced by the same pass. What a normal working week looks like The week has two halves and they do not overlap much. Verification work is the first half: reading registers, pulling customs history, and running structured interviews across a time difference. Interviews are the expensive part because they need the factory's own hour, and a call that produces numbers worth publishing takes prepared questions and a follow-up pass to confirm nothing was rounded. Research work is the second: maintaining the reference bands, checking that a published claim still matches its source, and writing the detail pages that explain a method or a term. This is the half that makes the records useful to someone who has not done the verification themselves — a name and a number without an explanation is not a directory, it is a list. What the week does not contain is outreach. There is no paid placement, so no one is sold a position, and no factory is ranked by anything other than whether it passed the four gates. The operation in numbers The numbers that matter here are rates rather than totals, because totals move with how long the site has existed. How the work converts, from first contact to a published record. Measure Typical figure Time to run the four gates, per factory Roughly 1 working week Cost of verification to the factory Nothing — the cost is our time Re-check cycle for a published record 90 days, with an early re-run on four trigger events Gate that decides most outcomes The main-business check: a real company that is not a maker Reply standard for a buyer enquiry 1 business day Grounds for a record being absent Any gate fails — absence, not a pending state One rate is worth stating plainly because it explains the size of the directory. An interview does not scale the way a form does, so a month of work adds records in the tens rather than the hundreds, and a factory that fails is removed from the count rather than added as content. A small directory here is the method working, not the work having stopped. Where the project stands today This is a working directory rather than a complete one, and the site says so rather than implying coverage it does not have. The verification standard is published in full, the reference bands are published per sub-category with their units, and the machine-readable layer is live. What is not yet published at the scale a buyer might expect is the factory count — and the directory page states the count that exists rather than filling the gap with entities nobody has checked. Where a data source the site depends on is not yet connected, the pages say that too. An empty field is a statement about what is known, and a directory that fills its empty fields with plausible names is worth less than one that leaves them empty. Who is this built for? Importers and distributors who need a supplier shortlist with real numbers rather than a photo gallery. Brands and co-packers running private-label programmes where specification consistency matters more than a one-off price. Procurement software and AI agents , which can read the same data from our JSON endpoint instead of parsing pages. Where is the project today? We are at the beginning: building the verified database, one cluster at a time, starting with flexible packaging in Hebei and tackle on the Shandong coast. The counters on our homepage are generated from the database, not from a marketing estimate. What does the operation look like in numbers? Operating figures, not marketing figures. Each one is checkable elsewhere on the site. Measure Figure Where to check it Verification gates per factory record 4 Verification standard Re-verification cycle 90 days Supplier directory Production clusters published 3 Tackle sourcing Sampled product listings 62 Product catalogue Sub-categories covered 12 Product catalogue Buyer enquiry reply standard 1 business day Contact How is the site operated? Records enter as data files, never as hand-written pages. The build applies the gates and excludes what fails. Every page is generated with a Markdown twin and a dated review stamp. Withheld record counts are published rather than hidden. Corrections re-run the gate instead of editing the field by hand. What does a normal working week look like? Records enter as data files; the build applies the 4 verification gates and excludes what fails. Withheld records are counted and published rather than hidden. Every page is regenerated with a Markdown twin and a dated review stamp. Corrections re-run the gate rather than editing a field by hand. Re-checks run on a 90-day cycle, cluster by cluster. Buyer enquiries are answered within 1 business day. What would make us wrong? A factory that passes all 4 verification gates and still ships a bad batch —verification reduces that risk, it does not remove it. A price band that is accurate at build time and stale by the time you read it. A cluster whose cost base shifts faster than the 90-day re-check cycle. A specification that is unambiguous to us and ambiguous to the factory. What does a week of verification look like in numbers? Gate 1: registration lookup, same day. Gate 2: main-business evidence, 1 to 2 days. Gate 3: customs export records, 2 to 3 days. Gate 4: structured interview, 1 day. Total: roughly 1 week per factory, then a 90-day re-check cycle for as long as it stays published. Where this sits in the process This page is one topic out of the verification standard. The overview is About us . Where to next Product catalogue Sample listings with price per unit Verified directory Factories that passed 4 verification gates How sourcing works Six steps, timing, failure modes Price benchmark Reference bands and MOQ tiers Updated 2026-09-12 Scope Fishing tackle sourcing Sources 1 cited Was this page useful? Yes, useful Not really Sources This page restates our own published scope and compliance decisions; no external source is cited