What I check
before a batch ships.

About the author

I am the technical director at Eastling Biotech in Xi'an. My work is the part of the business that decides what a batch is released against: the specification it has to meet, the certificate of analysis it ships with, and the answer to whatever technical question arrives alongside the quotation.

These guides are published under my review. They carry numbered references for a practical reason — almost every question we receive is a documentation question. What a purity figure actually covers, whether the counter-ion is stated, what a storage claim rests on. A documentation question deserves a document, not an adjective.

Where a number appears, it traces to a primary source listed at the foot of the page. Where a limit is our own release standard rather than a measured result for a specific lot, the text says so. If you find a sentence where that is not true, write to me and it gets corrected in place.

• Technical Director, Eastling Biotech Co., Ltd
• Specification drafting for 101 catalogue items
• Lot-level COA review before release

How a claim gets published

Every quantitative statement on this site passes the same five gates. They are deliberately boring, and they are the reason the articles carry numbered references instead of adjectives.

1. The claim has to be checkable. If a sentence contains a number, a range or a comparison, it must be traceable to a primary source: a peer-reviewed paper, a pharmacopoeial general chapter, or a regulatory guidance document.
2. The source has to be first-hand. Supplier websites, trade blogs, press releases and other marketing material are never cited as evidence — including our own. A specification page is an example of what we state, not proof of what is true.
3. The condition has to travel with the number. A stability figure without temperature, pH and time is not a fact, it is a rumour. We state the condition or we drop the figure.
4. The wording has to match the evidence. Where the literature is associative, we write “is associated with”, not “causes”. Where a limit is our own release standard rather than a measured result for a specific lot, we say so.
5. Someone else reads it. Drafts are checked against the lot documentation we actually hold before publication, and the review date is printed on the page.

What we will not publish

• Dosage, administration routes, cycles, or any implication of human or veterinary use. Every product on this site is sold for research and further manufacturing.
• Certifications we do not hold. If a document does not exist, the page says “documents on request” and lists only what can actually be issued.
• Comparative claims about named competitors. We describe what a good document looks like; the reader can compare.
• Numbers we cannot source. Where a figure would be useful but unevidenced, the sentence is rewritten or removed — not softened into vagueness.

Corrections

If a reference does not support the sentence it is attached to, or a figure is out of date, write to our documentation team with the article URL and the paragraph. Corrections are made in place, the revision date at the top of the article is updated, and the change is described in the reply. We do not quietly edit a published page.

Articles by Peter Song