Why most supplement ingredients don't work in a pouch (and how to screen one)

Every week a founder asks us to put a new ingredient in a pouch. Often it is a branded extract with a clinical study, a clean trademark, and a strong story on the supplier's sell sheet. Most of the time the honest answer is that it will not work the way the study says it works, and the reason has nothing to do with the ingredient's quality.

A pouch is a small, specific delivery format. It carries a few hundred milligrams of actives, it sits against the gum while saliva slowly carries most of its contents down the throat, and it has to taste acceptable for the whole wear time. Ingredients fail the format in two predictable ways: the studied dose is too big to fit, or the ingredient only works after gut transit and gains nothing from time in the mouth. This guide walks through both, shows the arithmetic, and ends with the screen we run before any formulation work starts.

PouchMade's standard pouch is 0.6g (600mg) total fill, with 0.5g and 0.7g formats available. The actives budget is typically 250-280mg per pouch, and about 300mg at most. Filler, flavor, sweetener, and carrier make up the rest, and that remainder is not optional: for moisture-sensitive actives, the cellulose carrier is also the buffer that protects shelf life.

That budget is shared. If a formula carries 50mg of caffeine and 100mg of L-theanine, everything else has to fit in the remaining 100-150mg. A branded extract that needs 200mg on its own leaves only 50-100mg for everything else, so it crowds out a caffeine-and-theanine stack rather than sitting beside it.

One more constant matters for dose math: a can holds 15 pouches. When an ingredient needs ten pouches a day to reach its studied dose, that is two-thirds of a can per day for one active, before anyone asks whether the flavor holds up.

The fastest screen is the studied dose. If the research measures the effective dose in grams, the pouch format cannot carry it, and no inclusion rate or clever blend fixes that. It helps to work one example all the way through.

Worked example: betaine. Anhydrous betaine (trimethylglycine) is sold in branded, beet-derived grades separated from sugar beet molasses. The beet origin leads some founders to assume it is a nitrate or nitric-oxide ingredient. It is not; a controlled trial found betaine does not move plasma nitrate or nitrite. The research doses are 1.5g per day for homocysteine support and about 2.5g per day for strength outcomes. At the 300mg ceiling, with nothing else in the pouch, 1.5g takes 5 pouches (1,500 / 300 = 5) and 2.5g takes 9 (2,500 / 300 = 8.3, rounded up). At a more typical 280mg budget, those become 6 and 9 pouches. A normal one-or-two-a-day habit delivers 300-600mg, which sits below 1g, a dose that has already been tested and did not produce the effect. Betaine is also a permanently charged molecule, so there is no route through the lining of the mouth, and it is deliquescent (it pulls in enough moisture to dissolve) above roughly 50% relative humidity. Every axis says no.

The same arithmetic applies across the sports and amino acid shelf:

Those pouch counts assume the ingredient takes the entire actives budget with nothing else in the formula. The market reflects the math: a wider scan of about 198 pouch products on major US marketplaces in September 2026 found none listing betaine, beta-alanine, creatine, L-citrulline, or glycine. That is not an untapped gap. It is the format saying no.

A small inclusion of a gram-dose ingredient can still have a job, for flavor, mouthfeel, or a label that names it honestly as a minor component. What it cannot do is carry a claim built on a gram-dose study.

The second failure is subtler, because the dose often fits. Many branded botanicals and polyphenols were studied as swallowed capsules, tablets, or beverages, and their mechanism depends on what happens after swallowing:

• Ginsenoside and astragaloside saponins (the actives in ginseng, notoginseng, and astragalus extracts) have low oral bioavailability even when swallowed and rely on gut bacteria to convert them into their active metabolites. Some branded astragalus and ginseng blends are marketed specifically to act on intestinal transporters, which is a gut mechanism by definition.
• Mushroom beta-glucans, the dominant studied fraction in reishi and turkey tail and much of chaga, act through immune contact in the gut wall. Their studied doses also run to grams: roughly 1.5-6g for reishi extract and about 3g for the best-studied turkey tail fractions.
• Chlorogenic acids, the headline compounds in coffee fruit extract, are only partly absorbed intact in the small intestine; much of the rest is metabolized by colon bacteria, with blood levels peaking hours after ingestion.
• Ellagitannins such as pomegranate punicalagins are barely absorbed intact and become active only after colon bacteria convert them to urolithins.
• Glucosinolates from broccoli and kale need the enzyme myrosinase, from crushed raw plant tissue or gut bacteria, to become isothiocyanates.
• Curcuminoids from turmeric are practically insoluble in water, poorly absorbed even from the gut, and stain oral tissue during a long hold.

Here is the part that is often stated wrong in both directions. A pouch does not block these ingredients. Most of what dissolves out of a pouch is carried by saliva and swallowed, so a pouch still delivers an oral, swallowed dose. The problem is that it delivers nothing more. Small, un-ionized molecules in the caffeine class can absorb partly through the oral mucosa; a chewing-gum study found caffeine reached peak blood levels faster than a capsule, though total exposure was the same or lower. Saponins, polysaccharides, polyphenol glycosides, amino acids, and minerals have no comparable route, and for almost every non-caffeine active there is little or no pouch-specific absorption data at all.

So for a gut-dependent ingredient, the pouch is a slow capsule with a taste problem. If the studied dose fits and the brand wants the format for its own reasons, that can be a legitimate product decision. But the evidence behind it is swallowed-dose evidence, the claims have to track that, and nothing about holding it under the lip makes it work better or faster. Saponins and polyphenols also tend to be bitter, astringent, or pigmented, and most branded extracts have no published taste or oral-hold stability data because they were designed for capsules and drinks.

The budget counts what goes into the blender, not the number on the marketing sheet. When an ingredient is sold as a standardized extract, the full extract weight counts against the 250-300mg, not the weight of the marker compound it is standardized to.

• A ginseng extract dosed at 50mg to deliver about 5mg of ginsenoside Rg1 uses 50mg of budget, not 5mg.
• The best-corroborated lion's mane dose in the research is 1,050mg per day of an extract standardized to erinacine A. That is 3.5 times the 300mg ceiling (1,050 / 300 = 3.5) before a single other active goes in.
• A typical rhodiola extract serving supplies only a few milligrams of salidroside, while a human trial of isolated salidroside used 60mg per day. Extract studies and isolated-compound studies do not transfer to each other.
• Carrier-loaded actives split across the budget. Alpha-GPC is commonly sold as a 50% powder as well as higher-purity grades; 60mg of alpha-GPC takes 120mg of pouch from the 50% material and about 67mg from a 90% grade (60 / 0.9 = 66.7).

Ask for the extract ratio, the marker percentage, and the carrier. If a supplier cannot tell you what the rest of the powder is, you cannot budget it.

The ingredients that work in a pouch share a profile: a studied dose in the tens of milligrams or lower, a well-characterized identity with a clear assay, and either a plausible oral-mucosal route or no need for one. The current market already clusters here. In a PouchMade review of about 36 non-nicotine pouch formulas on major US marketplaces in September 2026, caffeine appeared in 24, with disclosed doses of 22.5-200mg and 50mg the most common. Alpha-GPC appeared in 14 (disclosed doses 40-85mg), L-theanine in 13 (35-100mg), and citicoline in 3 (63.5-125mg).

• The caffeine class. Caffeine and theacrine are small molecules dosed in milligrams, and caffeine is the one active with published evidence of absorption through the mouth.
• Cholinergic and calming nootropics at market doses. L-theanine, alpha-GPC at typical market inclusions, and citicoline all fit comfortably beside caffeine.
• B vitamins. B12, B6, and methylfolate are dosed in micrograms to low milligrams and take almost none of the budget.
• Microgram actives. Huperzine A was disclosed at 0.03mg in the one formula that dosed it. At 30 micrograms in a 600mg pouch it is 0.005% of the fill, so the challenge is blend uniformity, not space.
• Standardized extracts at low milligram doses, where the studied dose is the extract weight and it is small. Ergothioneine at 25mg fits easily, for example, though its effects in the research build over weeks, so it belongs in a supporting role rather than as a felt hero ingredient.

These still need real formulation work. Alpha-GPC and several amino acids pull in moisture, bitter actives need masking, and microgram actives need pre-dilution and uniformity testing. But the problems are solvable, which is the point.

Run every candidate through these questions before paying for samples or packaging. Most ingredients that fail, fail at step one or two.

1. 1. Studied dose vs the budget. Find the dose used in the human research, not the supplier's suggested use level. Divide it by 250-300mg. If one pouch delivers under about a third of the studied dose, or the ingredient needs most of the budget on its own, it is not a hero ingredient.
2. 2. Extract weight vs marker weight. Budget the full extract and carrier weight. Confirm the extract ratio, marker percentage, and what the remainder of the powder is.
3. 3. Route of absorption. Was the evidence generated with a swallowed capsule or beverage? Does the mechanism depend on gut bacteria, intestinal transporters, or immune contact in the gut? If so, expect swallowed-dose performance at best, and write claims to match. Never claim faster or stronger absorption for a pouch without data for that ingredient.
4. 4. Evidence quality. Is the effective dose backed by an independent, indexed human trial, or only by a sponsor-funded study, an in vitro model, or animal data? Does the effect show up acutely, or only after weeks of daily use?
5. 5. Taste and handling. Bitterness, astringency, staining, and pigment all get worse over a 20-60 minute hold. Hygroscopic and deliquescent powders cake, discolor, and shorten shelf life. Ask for a taste sample of the raw powder, not the finished drink it was designed for.
6. 6. Regulatory status. Is it a legal dietary ingredient in the US? Is there an NDI notification or a clear pre-1994 history? FDA has objected to at least one NDI notification on the ground that the substance, L-BAIBA, was not a dietary ingredient at all, and FDA's position is that synthetic copies of botanical constituents are not dietary ingredients, which leaves nature-identical versions of botanical compounds on uncertain footing. A GRAS notice for a food use does not automatically cover a pouch. Review anything novel with regulatory counsel; see our overview of how pouches are regulated.
7. 7. Testing. Can the ingredient be identified and assayed with a real method, typically HPLC for markers? Does the certificate of analysis show identity, marker assay, heavy metals, and microbial limits? For microgram actives, can the finished blend be tested for content uniformity?

An ingredient that passes all seven is a candidate. One that fails the first three is a marketing story looking for a format.

PouchMade manufactures oral pouches as dietary supplements under 21 CFR 111 in an FDA-registered, GMP-certified facility in Dallas, Texas. We would rather tell you an ingredient does not fit at the brief stage than after you have printed cans. When a requested ingredient fails the screen, we look for the actives that deliver the positioning you want inside the budget, and build the stack through custom formulation.

If you are weighing a branded ingredient for a nootropic, energy, or functional pouch, send us the ingredient and its studied dose through our project intake. We will run it through the same arithmetic and tell you plainly whether it can do its job in a pouch. For the broader production picture, start with Pouch manufacturing 101.