On this page
Key takeaways
● Third-party Certificate of Analysis from an ISO/IEC 17025-accredited independent laboratory — batch-specific, heavy metals (lead, cadmium, mercury, arsenic). Single strongest quality signal.
● Peptide molecular weight disclosure — ideally 2–3 kDa average with published range. Lower absorbs somewhat better.
● Source transparency: bovine (grass-fed disclosed), marine (species and sourcing), or porcine. All work; disclosure matters.
● Type I and III for standard skin/joint/bone support. Type II is a separate category (UC-II ~40 mg).
● Cost per gram at your target dose — not price per tub. Calculate at your indication-anchored dose.
● Skip the marketing distractions: flavourings, added biotin at meaningless doses, "multi-collagen" blends without evidence, celebrity endorsements.
Quick answer
Choose collagen products based on criteria, not brand marketing. In decreasing order of importance: (1) batch-specific Certificate of Analysis from an ISO/IEC 17025-accredited independent laboratory covering heavy metals; (2) disclosed molecular weight distribution — ideally 2–3 kDa average with published range; (3) transparent source (bovine grass-fed disclosed, marine species and sourcing disclosed, or porcine); (4) Type I and III collagen for standard skin/joint/bone/tendon support; (5) cost per gram at your indication-anchored target dose. The right brand is the one that transparently meets these criteria at a defensible price. Learn the criteria and any product becomes evaluable.
Why a criteria-first approach beats brand rankings
Consumer supplement categories have short product half-lives. A brand that was the best-tested option in 2023 may be reformulated in 2024 or acquired and cost-optimised in 2025. A brand that had an excellent Certificate of Analysis in one batch may fall short in another. Rankings age poorly, and rankings from sources with affiliate relationships age with additional bias.
The alternative is to learn the criteria that matter, verify a product against those criteria at the point of purchase, and re-verify periodically. This puts the reader in the position of doing five minutes of consumer diligence rather than trusting a static ranking. The tradeoff is more effort per purchase; the upside is durable independence from the marketing cycle. This guide teaches the criteria; the rest is on you.
Criterion 1 — Third-party Certificate of Analysis (the single most important)
Collagen is a category where independent third-party testing matters. Independent testing programmes have found measurable lead, cadmium, mercury, and arsenic contamination in some collagen brands — not universally, but often enough that the category-level concern is real, not niche (5). Marine collagen has slightly higher theoretical mercury risk from fish source; bovine has different but non-zero heavy metals profile. Neither source is automatically clean; both benefit from testing verification.
What to look for:
● A batch-specific Certificate of Analysis — the specific batch you are buying (or planning to buy). Blanket "our products are tested" claims are less informative than a document tied to the specific manufacturing lot.
● Testing performed by an ISO/IEC 17025-accredited independent laboratory — this is the international standard for testing laboratory competence. Named accreditation matters.
● Testing covering the heavy metals of concern — lead, cadmium, mercury, and arsenic at minimum. Bonus points for microbial testing, allergen testing, and identity confirmation.
● Publicly accessible or on request — a company that publishes CoAs on the product page or provides them promptly on request is signalling confidence in what testing reveals.
What is not enough:
● "Third-party tested" as an unspecified marketing claim without documentation.
● Testing by an in-house laboratory (not independent).
● Testing that only covers identity confirmation ("yes, this is collagen") without heavy metals.
● Certifications by industry-funded rather than independent bodies.
Criterion 2 — Molecular weight disclosure
The bioactive peptides that drive collagen's clinical effects are small — Pro-Hyp and Hyp-Gly at approximately 200–300 daltons individually. But the whole peptide mixture in a commercial product spans a molecular weight range, and where that range sits affects absorption and clinical signal strength.
Commercial hydrolysed collagen typically falls in the 2–6 kDa range. Peptides in the 2–3 kDa portion of the range appear to absorb somewhat better and produce slightly stronger clinical signals in some trials than heavier peptides at 5–6 kDa (3). This is not a dramatic difference — it will not turn a placebo effect into a transformative one — but at the margin, lower-MW peptides are better.
What to look for:
● Published average molecular weight — ideally displayed on the product page or on request. "Average MW <3,000 Da" is a positive signal.
● Published molecular weight distribution or range — even better, though rarer.
● Language that signals engineering: "low molecular weight peptides," "bioactive collagen peptides," "optimised for absorption." These claims are marketing-adjacent and worth verifying against actual MW data.
What is not enough: silent labels that provide neither MW nor distribution data. The absence of this information is not disqualifying — many perfectly acceptable products do not publish MW — but its presence is a positive signal about manufacturer transparency.
Criterion 3 — Source transparency
Marine, bovine, and porcine collagen all work comparably in trials. The 2023 Pu meta-analysis found no significant source difference on skin outcomes (2). What matters is that the source is transparently disclosed and that peripheral considerations (welfare, sustainability, contamination profile) are addressed appropriately.
What to look for:
● Species and origin disclosed — "marine collagen from wild-caught North Atlantic cod" is more informative than "marine collagen".
● Welfare or sustainability certification where applicable — MSC certification for marine, grass-fed or regenerative certification for bovine.
● Traceability language — the ability to trace back to specific farms, fisheries, or regions. Common in higher-end products; not universal.
● Compatibility with your dietary preferences — kosher certification, halal certification, avoidance of species you exclude.
For the full source-comparison analysis, see our marine vs bovine article.
Criterion 4 — Type I and III for standard support (not Type II)
Almost all hydrolysed collagen supplements are Type I and Type III — the collagen types of skin, tendon, ligament, and bone matrix. This is what supplement trials use for skin, joint, and bone indications. Marine and bovine sources both deliver Type I and III appropriately.
Type II collagen is a separate product category with a completely different mechanism. Undenatured Type II collagen (UC-II) works via oral tolerance induction at Peyer's patches at approximately 40 mg per day — three orders of magnitude below hydrolysed collagen doses. Type II supplements are legitimate for isolated knee osteoarthritis pain but are not interchangeable with hydrolysed Type I/III collagen for combined skin, bone, and joint goals. Do not confuse the two categories at purchase. See our hydrolysed vs non-hydrolysed article for the mechanism distinction.
"Multi-collagen" blends claiming to include Type I, II, III, V, and X are common. Trial evidence for these blends is essentially the sum of individual constituent evidence — there is no published RCT demonstrating that the blend outperforms a single-source Type I/III product on any outcome. The main benefit is convenience or marketing appeal, not efficacy.
Criterion 5 — Cost per gram at your target dose
The relevant comparison is not price per tub but price per gram delivered at your indication-anchored daily dose. Some products with expensive per-tub pricing become reasonable per-gram; some cheap tubs become expensive because the effective daily dose requires an oversized serving.
Practical calculation:
● Identify your indication-anchored dose from the dosage article. E.g., 5 g/day for postmenopausal bone; 10 g/day for skin or knee OA.
● Convert tub size to daily servings at your dose (a 300 g tub at 5 g/day = 60 days; at 10 g/day = 30 days).
● Divide tub cost by number of days at your dose to get cost per day.
● Multiply by 30 for monthly cost, or 365 for annual cost, at your dose.
Typical market prices for reasonable-quality hydrolysed collagen at 10 g/day translate to £15–£40 monthly. Above £40/month, verify that the extra spend is buying real quality differentiation (transparent testing, superior MW, premium sourcing) rather than marketing. Below £15/month, extra scrutiny on Certificate of Analysis is warranted — reputable third-party testing costs money.
What to ignore or de-prioritise
● Flavourings. Vanilla, chocolate, berry — makes it easier to consume but does not affect efficacy. Watch for added sugar in flavoured products. Unflavoured is often the best value.
● Added biotin at low doses. Meaningful biotin doses for hair and nails are 2,500–5,000 mcg/day. A collagen product with 30 mcg of biotin is a marketing gesture, not a clinically relevant biotin dose.
● Added vitamin C at low doses. If your dietary vitamin C is adequate, added vitamin C in a collagen product is unnecessary. If it is inadequate, address that separately with a proper vitamin C intake plan.
● Added "beauty" ingredients (biotin, hyaluronic acid, silica, various herbs) at trace doses. If HA is a serious goal, take it as a standalone supplement at 120–200 mg/day; trace amounts in blended products are marketing, not clinical dose.
● Celebrity endorsements and influencer partnerships. Adds nothing to product quality. If anything, worth extra scepticism about pricing markup.
● "Multi-collagen" and "5 types of collagen" claims. Marketing, no trial-outcome differentiation.
● Grass-fed labelling as an efficacy claim. Grass-fed matters for welfare and environmental considerations, not for the collagen molecule itself. Do not pay a premium for grass-fed if efficacy is your criterion.
A worked example of the decision
A reader wants collagen for combined skin and joint outcomes. Their daily dose is 10 g. They are comparing two products at a shop:
● Product A: £25 for a 300 g tub (30 days at 10 g/day = £25/month). Bovine, grass-fed disclosed. Published average MW 2.5 kDa. Third-party CoA available on the website, ISO 17025 lab named, batch number matches. Unflavoured.
● Product B: £45 for a 300 g tub (£45/month). Marine, species and origin disclosed. Published MW not specified. "Third-party tested" claim with no CoA link. Contains added biotin (50 mcg), added vitamin C (10 mg), and "beauty blend" of various herbs.
Analysis: Product A is the better choice. Lower cost, transparent CoA with named ISO lab, disclosed MW. Product B costs 80% more for less transparency and marketing extras ("beauty blend", trace biotin, trace vitamin C) that add no real clinical value. The marine source of B is not obviously an advantage — meta-analysis shows no source difference. The reader's £20/month savings could go toward a proper standalone HA supplement at 200 mg/day if skin hydration is a priority.
What we still don't know
● Whether industry certification programmes (USP-Verified, NSF Certified for Sport, informed-sport) meaningfully differentiate product quality at the level that matters for clinical outcomes.
● Whether specific manufacturers' bioactive peptide profiles — e.g. specific Pro-Hyp:Hyp-Gly ratios — matter more than average molecular weight.
● Whether third-party testing frequency (per-batch vs periodic) affects real-world contamination exposure.
Bottom line
Choose hydrolysed collagen by criteria, not by brand ranking. In decreasing order of importance: batch-specific Certificate of Analysis from an ISO/IEC 17025-accredited independent laboratory covering heavy metals; disclosed molecular weight (ideally 2–3 kDa average); transparent source; Type I and III collagen for standard indications; cost per gram at your target dose. Ignore flavourings, low-dose added ingredients, celebrity endorsements, and multi-collagen marketing. A well-verified budget-tier product at £20/month beats a poorly-verified premium-tier product at £45/month. Learn the criteria once and evaluate any product yourself. See our pillar guide for the complete picture.
Frequently asked questions
What is the best hydrolysed collagen brand?
There is no single "best" brand — brand offerings change, and the criteria that matter (third-party testing, molecular weight, source, cost per gram) are what should drive your specific product choice. Learn the criteria in this article and evaluate any product against them.
How do I know if a collagen product is high quality?
Verify: (1) batch-specific Certificate of Analysis from an ISO/IEC 17025-accredited independent laboratory; (2) disclosed molecular weight (ideally 2–3 kDa); (3) transparent source with species and origin named; (4) reasonable cost per gram at your target dose.
Is expensive collagen actually better?
Not necessarily. Higher price sometimes reflects better testing, superior MW, or premium sourcing — but often it reflects marketing markup, celebrity endorsement costs, or unnecessary added ingredients. Verify what you are actually paying for.
Should I buy multi-collagen or single-type collagen?
Single-type (Type I and III) has the trial evidence base for standard skin, joint, and bone indications. Multi-collagen blends have no published trial evidence showing they outperform single-type. Do not pay a premium for multi-collagen unless the product has independent quality advantages.
Is grass-fed collagen better than conventional?
Not for efficacy. Grass-fed matters for welfare and environmental considerations. The collagen molecule extracted is chemically identical. Do not pay a premium for grass-fed if efficacy is your criterion; do pay if welfare or environmental values matter to you.
How much should I spend on collagen per month?
Typical reasonable market range at 10 g/day is £15–£40 monthly. Below £15 warrants extra scrutiny on quality documentation; above £40 warrants scrutiny on what the premium is actually buying.
What is the difference between USP-Verified and NSF Certified?
Both are independent verification programmes. USP-Verified confirms identity, potency, purity, and manufacturing quality (4). NSF Certified for Sport additionally screens for banned athletic substances. Either is a positive signal; neither is essential if a batch-specific ISO 17025 CoA is provided directly.
References
1. León-López A, Morales-Peñaloza A, Martínez-Juárez VM, et al.. Hydrolyzed collagen — sources and applications. Molecules 2019. https://pmc.ncbi.nlm.nih.gov/articles/PMC6891674/
2. Pu SY, Huang YL, Pu CM, et al.. Effects of oral collagen for skin anti-aging: a systematic review and meta-analysis. Nutrients 2023. https://doi.org/10.3390/nu15092080
3. Iwai K, Hasegawa T, Taguchi Y, et al.. Identification of food-derived collagen peptides in human blood after oral ingestion of gelatin hydrolysates. J Agric Food Chem 2005. https://pubmed.ncbi.nlm.nih.gov/16076145/
4. US Pharmacopeia. General guidelines for supplement testing standards. USP 2023. https://www.usp.org/verification-services/verification-guidelines
5. ConsumerLab. Product review — collagen supplements: heavy metals, potency, and labelling. ConsumerLab 2024. https://www.consumerlab.com/reviews/collagen-supplements-review/collagen/