If your goal is muscle, buy whey. Collagen carries zero tryptophan, is low in leucine, and its DIAAS is effectively zero. In a 2023 trial with muscle biopsies, 30 g of collagen after training did not raise muscle connective protein synthesis above placebo. TARE sells collagen and does not sell whey.
That is the whole answer, and it is worth being clear about who is giving it. This page is published by a company whose one powder in this comparison is the collagen. The recommendation still goes the other way, because the question in the search box is about muscle and the evidence on muscle is not ambiguous.
What follows is why: the amino acid arithmetic, the four trials that measured muscle directly, and then the narrower thing collagen is actually for, with its own contradiction printed alongside it.
Collagen or whey: which one should you buy?
Whey, if the goal is muscle. It is not close and it is not contested. Collagen carries zero tryptophan, is low in leucine, and its DIAAS is effectively zero. Whey supplies all nine, including the one collagen is missing entirely. TARE sells collagen, does not sell whey, and would rather write that sentence than sell somebody the wrong tub.
| Property | Collagen | Whey |
|---|---|---|
| Tryptophan | None | Present |
| Leucine | Low | High |
| DIAAS | Effectively zero | High |
| Complete protein | No | Yes |
| Raised muscle protein synthesis in tracer trials | No | Yes |
| What it is bought for here | Tendon, under load | Muscle |
| Sold by TARE | Yes | No |
Nothing in that table is a close call, and the last two rows are the reason the page exists. These are not two versions of one product competing on quality. They are two different proteins doing two different jobs, and putting the collagen in the whey’s place is a trade down.
What is the actual amino acid difference?
Collagen is about a third glycine and contains no tryptophan at all. Alcock and colleagues, Front Nutr 2019;6:163, fed 20 g of each and measured what reached the blood: collagen produced a plasma glycine peak of 683 against 260 µmol/L for dairy protein, and dairy produced the higher leucine peak, 267 against 189. Two proteins delivering two different things.
Leucine is the amino acid that switches on muscle protein synthesis, and it is the one collagen is short of. Tryptophan is one of the nine a human body cannot make, and collagen has none of it. Those two facts are the whole of the protein-quality argument and there is no version of them TARE can argue with.
DIAAS effectively zero is the formal record of it. The score works by finding the single most limiting indispensable amino acid in a protein and grading the whole protein by that one. When the limiting amino acid is present at zero, the score is zero, no matter how much nitrogen the rest of the powder contains.
That is also why the plasma-glycine number above is a composition fact and not a benefit. It says what arrives in the blood. It does not say what the body does with it, and this page is not going to pretend the second follows from the first.
What happened when a laboratory measured muscle directly?
Aussieker and van Loon, Med Sci Sports Exerc 2023;55(10):1792–1802, infused a labelled amino acid and took muscle biopsies from 45 people. Thirty grams of whey after a resistance session raised myofibrillar protein synthesis above placebo. Thirty grams of collagen did not raise muscle connective protein synthesis above placebo. The same laboratory repeated it in 2025 in men and women, against a control of the identical amino acids in free form, and nothing moved there either.
That design is the strongest one available for this question, because it does not infer anything. A labelled amino acid goes into the bloodstream and the muscle itself is sampled with a needle, so the number is a synthesis rate rather than a scan of a limb. Four trials, from two laboratories, point the same way.
- 2023, n = 45, recreational athletes. Whey raised myofibrillar synthesis. Collagen did not beat placebo on connective protein synthesis.
- 2025, n = 45, men and women. Thirty grams of collagen, and the same amino acids given in free form, both failed to raise myofibrillar or connective synthesis in the rested or the exercised leg. Baar, an author of the 2017 gelatin study this whole category rests on, is a co-author here. Med Sci Sports Exerc 2025;57(11):2394–2408.
- 2020, n = 22, older women, six days at 60 g a day. Whey increased muscle protein synthesis. Collagen did not raise the longer-term measure above baseline in either leg. Am J Clin Nutr 2020;111(3):708–718.
- 2024, n = 31, men over 72, 50 g a day added to a controlled diet. Whey worked. Pea worked. Collagen did not move it. Am J Clin Nutr 2024;120(1):34–46.
The last one is the sentence that settles it. Collagen did not merely lose to whey, which is the protein everyone expects to win. It lost to pea, in a design where the only thing that changed was which powder went into the same diet.
Is collagen a complete protein?
No. A complete protein supplies all nine indispensable amino acids in usable amounts, and collagen supplies zero tryptophan, which alone disqualifies it. A DIAAS of effectively zero is the formal way of recording the same thing. The 9 g of protein declared on our own panel is a true number answering a question nobody should be asking of this powder.
It is worth sitting with that for a sentence, because the 9 g is printed on the tub and it is not a trick. Protein on a Supplement Facts panel is measured by nitrogen, and collagen is full of nitrogen. The panel is not lying. It is measuring quantity in a place where the reader is assuming quality.
That is the single most useful thing to carry off this page and onto anyone else’s container. A protein gram on a panel is a nitrogen result. Whether those grams can build anything is a separate question the panel does not answer and is not designed to.
Can collagen replace a protein powder?
No, and the trials that tested the substitution are why. Over 10 weeks of supervised training, whey produced greater increases in muscle thickness than leucine-matched collagen — Jacinto and colleagues, Int J Sport Nutr Exerc Metab 2022;32(3):133–143. In men over 72 given 50 g a day of extra protein, whey worked, pea worked, and collagen did not move muscle protein synthesis at all — McKendry, Am J Clin Nutr 2024;120(1):34–46.
The 10-week trial is the one worth reading twice, because its designers tried to fix collagen’s problem before they tested it. They added free leucine to the collagen arm until the two powders matched on the amino acid collagen is short of. Whey still produced more muscle thickness — 8.4% against 5.6% in the vastus lateralis, 10.1% against 6.0% in the biceps. Twenty-two untrained adults, eleven per group.
And the second half of that same trial belongs here too, because it is a negative for both powders: strength and power rose the same amount in both groups. That is consistent with everything else in this literature, and it is why no strength claim appears anywhere on this site.
So the substitution question has a plain answer. If a reader is buying protein powder to support training, swapping it for collagen is a trade down, and no dosing change fixes it. The full evidence review is in its own article, including the parts that go against our own product.
So what is collagen actually for?
Tendon cross-sectional area and stiffness, at 15 to 30 g a day, in someone already training hard. That is the finding the 2026 systematic review in J Funct Morphol Kinesiol grades strong, and the same review grades the evidence strong against any effect on muscle strength. It has a contradiction attached: the trial that ran 15 g for 15 weeks and imaged the tendon by MRI found no difference at all.
The review ran to eight randomised trials and 257 people, and every one of those trials paired the powder with resistance or plyometric training. The training is the larger half of the intervention, which is why the dose on this site is tied to a session rather than to a day.
The contradiction is not buried here either. Balshaw and colleagues, Med Sci Sports Exerc 2023;55(11):2083–2095, ran 39 men on 15 g a day for 15 weeks and imaged the patellar tendon by MRI: no between-group difference on any tendon measure. That trial is larger than most of the favourable ones, longer than most, and used the harder instrument.
One more thing belongs in that paragraph because it is against our own interest. The two trials that separated from placebo on tendon stiffness both dosed 30 g, and the trial that dosed 15 g separated on nothing. Fifteen grams is the figure this site recommends, taken from the 2017 trial that set the timing, and we did not go to 30 g. Those trials are named, with their sample sizes, in the evidence article.
Which is the honest shape of this product. It is a narrow intervention with a real mechanism, a small and mostly male evidence base, and one large trial that says no. Whey is not in that position on its own question, and pretending otherwise would be the easiest sentence on this page to write and the least defensible.
Can you take collagen and whey together?
They do not compete for the same slot. Collagen goes in 30 to 60 minutes before the session that loads the tendon, with 50 to 100 mg of vitamin C. Whey is a daily protein-intake decision. The one trial that put both in the same drink — a 25 g whey and 5 g collagen blend, Med Sci Sports Exerc 2025;57(3):544–554 — raised myofibrillar synthesis, which whey alone already does, and its connective-tissue result failed in the exercised leg. A co-author is affiliated to Gelita AG, a collagen manufacturer. That is not a case for a blend.
Read that trial the way its own numbers read. The drink was 83% whey. Whey on its own already raises myofibrillar synthesis, so the headline result is attributable to the ingredient nobody was testing. The connective-tissue result — the one the collagen was in there for — reached significance in the resting leg and not in the exercised leg, which is the leg that matters.
So the practical answer is the boring one: keep them separate, keep them at different times, and do not pay extra for a pre-mixed version of two powders that are cheaper apart. The timing detail for collagen is worked through in how much collagen for tendons.
What is on the TARE panel, and what was measured?
Under Load declares 10 g per serving: 9 g protein, 35 calories, 0 fat, 0 carbohydrate, two 5 g scoops, 35 servings, one ingredient. The only measured protein result in existence is 9.8 g per 10 g serving, against a 9 g specification, from ViaCore Nutrition — the manufacturer’s own laboratory — on lot 121125450. At the 15 g protocol dose that is 1.5 servings, which is 14.7 g by arithmetic and was never measured.
Both halves of that sentence are deliberate. The 9.8 g is a laboratory result on a real 350 g lot of our own product, dated 12/22/2025 by ISO 14891:2002, and it came back above its own specification. The 14.7 g is multiplication. Only one of those two numbers was ever measured, and it is not the larger one.
The laboratory is worth naming rather than footnoting. ViaCore Nutrition, El Reno, Oklahoma, is the company that makes the powder, testing the powder it made. That is a weaker check than an outside laboratory, and a reader deserves to hear it from us rather than work it out. Everything the document does and does not cover is transcribed row by row on Testing & Certification.
And the 15 g is a directions statement, not a serving. The declared serving is 10 g, two scoops, 35 servings, full stop. At three scoops a session the tub holds 23 whole doses rather than 35, which is the arithmetic no seller has a reason to print about its own container and which is printed here anyway.
What we sell, and what we do not
Under Load is collagen Type I and III, bovine, one ingredient, 350 g, with a 5 g scoop. It is a pre-training dose for tissue under load. It is not a protein supplement, it is not sold as one, and the amino acid profile above is only an indictment of the product it is not.
There is no whey in this range and there is not going to be one for the sake of this article. A reader whose question is muscle is better served by a good whey than by anything on this site, and saying so costs a sale we were never entitled to. Selling alongside a protein powder is a larger market than trying to sell instead of one, and it happens to be the truthful position as well.
The same instinct produced the rest of this range. We audited our supplier and two products did not survive it; they were retired rather than quietly restocked, and the reasoning was published rather than buried. That is the only reason a page as claims-light as this one is worth anything.
More in this cluster: does collagen actually work, in full, how much collagen for tendons, the same substitution question asked about colostrum, and what a certificate of analysis proves, and what ours do not. Or read the documents themselves on Testing & Certification.