In this LAB note
Question · Hypothesis · Method · Results · What Failed · What We Learned · What Changed at MOMS
Question
At the doses that actually appear in our recipe database, is turmeric a functional ingredient or a label ingredient - and does the Score gap between turmeric and non-turmeric recipes belong to the turmeric at all?
Hypothesis
We expected the dose to be too small to matter physiologically, and we expected the data to say so - turmeric and non-turmeric recipes scoring roughly the same, with the spice earning its place on aroma and color rather than biology. What we did not expect was for the data to appear to defend turmeric, and for that defense to fall apart on inspection.
Method
Two published lines set the dose arithmetic. Tayyem and colleagues (Nutrition and Cancer, 2006) measured curcumin by HPLC across commercial turmeric and curry powders and found that pure turmeric powder averaged 3.14% curcumin by weight - the conversion factor that turns our grams into milligrams of active compound. Shoba and colleagues (Planta Medica, 1998) supplied the ceiling on the other end: 2,000 mg of curcumin taken alone produced serum concentrations in human volunteers that were undetectable or extremely low, and only co-administration of 20 mg of piperine raised bioavailability meaningfully. Internally, we parsed the Functional column of all 201 recipes in the formulation database - 19 distinct functional ingredients across 179 placements - recorded every dose in grams, and compared Score three ways: turmeric versus no turmeric, turmeric alone versus turmeric alongside a second functional ingredient, and Score against the number and total gram weight of functional ingredients regardless of identity.
Results
The first cut flattered turmeric. The 31 recipes containing it average a Score of 15.71; the 170 without it average 14.85 - a gap of 0.86. Then the dose arithmetic. The median turmeric dose in the database is 0.5g, and 24 of the 31 recipes sit below 1g. At Tayyem's 3.14%, 0.2g of turmeric delivers about 6 mg of curcumin and 0.5g delivers about 16 mg - against the 2,000 mg that Shoba's volunteers absorbed almost none of. Our most heavily dosed recipe, WAY-W1-HYDRATE-007-V1 (Steady Waters), carries 7.5g of turmeric for roughly 236 mg of curcumin, still an order of magnitude below a studied dose, and scores 15 - a point below the turmeric group mean. So the gap belongs somewhere else. Turmeric recipes carry 2.0 functional ingredients on average against 0.9 across the database, and functional ingredient count tracks Score on its own: recipes with no functional ingredient average 14.57 (n=77), one averages 15.05 (n=78), two or more average 15.54 (n=46). Split the turmeric set by that variable and it separates cleanly - turmeric as the sole functional ingredient averages 15.11 (n=9), while turmeric alongside a second averages 15.95 (n=22). Total functional gram weight correlates with Score at r = +0.240, grams of nut and seed specifically at r = +0.199, and turmeric dose within the turmeric set at only r = +0.102. EY-W1-ENERGY-001-V1 (Sunrise Surge) makes it concrete: Turmeric 0.2g + Cashew 5g, 393.5 kcal, 34.5g protein, 11.7g fat, 9.4g fiber, Score 16. The cashew is twenty-five times the mass of the turmeric. The turmeric was riding along.
What Failed
The two recipes whose Goal literally names the mechanism. EI-W4-RENEW-005-V1 (Renew & Anti-Inflammatory) and EI-W4-RENEW-006-V1 (Soothing Renewal) are both built for Renew & Anti-Inflammatory, and both carry Turmeric 0.5g as their only functional ingredient - roughly 16 mg of curcumin each. Both score 13, the joint lowest in the entire turmeric set. The ingredient we were leaning on to carry an anti-inflammatory claim was present at a dose that cannot support one, and the recipes around it were thin: Renew & Ease runs 303.1 kcal with 6.4g fat and 6.4g fiber. Placement failed too. Turmeric is scattered across almost twenty different Goals and appears in only two of the Renew & Anti-Inflammatory recipes - it was being added by habit, not by intent. Our attempted rescue failed as well. We checked whether the 11 recipes whose only functional ingredient is a sub-gram spice were quietly fine, and their Scores run from 12 to 18 with a mean of 15.18 - indistinguishable from the database, which is the point. The spice is not predicting anything in either direction.
What We Learned
Dose is the ingredient. Contains turmeric is a fact about a label, not a fact about a formula, and our Score was never rewarding the spice - it was rewarding the cashew, chia, or pecan that happened to be listed next to it. Curcumin is also fat-soluble and poorly absorbed on its own, which makes gram weight and the fat it travels with inseparable questions. A sub-gram spice in a 380-calorie shake is a flavor and color decision. That is a legitimate reason to include it. It is not a reason to call it functional.
What Changed at MOMS
We split the Functional column in two. Ingredients present at a material dose - nuts, seeds, and nut butters at 5g or more - remain functional and continue to count toward a recipe's formulation credit. Sub-gram spices are now recorded as flavor and color, and no recipe can satisfy its functional slot with a spice alone; the 11 recipes currently doing so are queued for reformulation. Where a Goal makes an anti-inflammatory claim, turmeric must be dosed at 3g or more and must travel with at least 8g of fat in the same recipe, since curcumin does not absorb without one. And we stopped crediting turmeric for a recipe's Score at all - the number we defend now is total functional gram weight, which is what the data was measuring the whole time.
References
1. Tayyem RF, Heath DD, Al-Delaimy WK, Rock CL. Curcumin content of turmeric and curry powders. Nutrition and Cancer. 2006;55(2):126-131. 2. Shoba G, Joy D, Joseph T, Majeed M, Rajendran R, Srinivas PSSR. Influence of piperine on the pharmacokinetics of curcumin in animals and human volunteers. Planta Medica. 1998;64(4):353-356.
Keep reading
Previous experiment
Next experiment
Not sure which shake fits you? We match the formulation to your body type.
Find your body type