In this LAB note
Question · Hypothesis · Method · Results · What Failed · What We Learned · What Changed at MOMS
Question
We call one slot in every recipe the Functional slot, and we named it for the actives: turmeric, ginger, cinnamon, cacao. Across 201 recipes that slot holds 935 grams of ingredients. How many of those grams are actually the actives, and how many are nuts and seeds?
Hypothesis
We expected the Functional slot to be mostly botanicals. Turmeric is the single most frequent entry in it, and internally "functional" has always been shorthand for the small-dose actives rather than for fat. Nuts and seeds, we assumed, appeared occasionally for texture or for omega-3s and would be a minority of the slot's mass. We also expected that where we had placed a nut or a seed, we had placed it deliberately — next to the carotenoid-dense produce that needs fat to be absorbed. If the slot was doing quiet absorption work, our own recipes should show the pairing.
Method
We pulled all 201 recipes from the MOMS formulation database and parsed the Functional column of each into individual ingredients with their gram weights. We sorted the entries into two groups: lipid-dense whole foods (cashew, chia seeds, flaxseed, pecan, walnut, almonds, peanut butter) and everything else (turmeric, ginger, cinnamon, lemon, cacao, cocoa, seaweed, tofu, agave, strawberry, maple syrup). We summed the mass of each group, estimated the fat each lipid entry contributes using standard composition values — cashew 43.8 g fat per 100 g, chia 30.7, flaxseed 42.2, pecan 71.9, walnut 65.2 — and expressed that as a share of the recipe's finished fat. We then tested whether lipid-carrying Functional slots are preferentially placed in recipes whose Fruit and Vegetable or Body slot is carotenoid-dense. For the absorption question we used two human crossover trials. Brown and colleagues (2004) fed seven subjects three otherwise identical salads of spinach, romaine, cherry tomato and carrot with dressings containing 0, 6 or 28 grams of canola oil, and measured carotenoid appearance in plasma chylomicrons by HPLC with coulometric array detection. Unlu and colleagues (2005) added 150 g of avocado or 24 g of avocado oil to salad in eleven subjects and measured alpha-carotene, beta-carotene and lutein absorption against an avocado-free control.
Results
The Functional slot is small. Across all 201 recipes it averages 4.7 grams — the second-lightest slot in the shake, ahead only of Flavor at 4.1 grams, against 229.2 grams of liquid, 223.6 grams of protein, 92.3 grams of fruit and vegetable and 41.6 grams of body. Only 124 of the 201 recipes carry a Functional slot at all. Of the 935.1 grams sitting in that column across the whole database, 727.6 grams — 77.8 percent — are cashew, chia seeds, flaxseed, pecan, walnut, almonds or peanut butter. Eighty-one recipes contain one of them. The botanicals that gave the slot its name are trace by comparison: median turmeric dose 0.5 g, median cinnamon 0.2 g, median ginger 2.5 g, against a median cashew of 7.6 g, chia of 6.3 g and pecan of 9.0 g. EY-W4-CALM-001-V1 ("Quiet Recovery", Standard, Score 20) states it in one line. Its Functional slot is Turmeric 0.5 g + Cashew 15 g. The cashew outweighs the turmeric thirty to one and supplies roughly 6.6 g of the shake's 10.0 g of total fat, about 66 percent of it. EI-W1-CLEANSE-017-V1 ("Slow Clean") has no botanical at all: its Functional slot is Cashew 18 g, contributing about 7.9 g of 11.5 g total fat. Across all 81 lipid-carrying recipes the Functional slot supplies a median 32 percent of the finished shake's fat, and the group averages follow — recipes with no Functional slot average 7.8 g of fat, those with a botanical-only slot 10.0 g, those carrying a nut or seed 12.6 g. That range sits directly on the dose Brown and colleagues tested. Fifteen grams of cashew delivers about 6.6 g of fat — essentially their 6-gram arm, where carotenoid appearance in chylomicrons moved from negligible with fat-free dressing to significantly higher (P less than 0.04), and rose further again at 28 grams (P less than 0.02). Unlu and colleagues saw alpha-carotene, beta-carotene and lutein absorption reach 7.2, 15.3 and 5.1 times the fat-free control after 150 g of avocado or 24 g of avocado oil. "Quiet Recovery" pairs its 15 g of cashew with 70 g of pumpkin, and 56 of the 81 lipid-carrying recipes contain carotenoid-dense produce of some kind.
What Failed
The pairing is not a design, and the number that looked strongest is the one that fails. Fifty-six of 81 is 69 percent — but the base rate in every other recipe is 77 of 120, or 64 percent. A five-point gap on samples this size is indistinguishable from chance. We placed nuts and seeds next to carotenoid-dense produce at roughly the rate we would have by accident. Worse, 77 recipes carry carotenoid-dense produce with no nut or seed in the Functional slot at all, and 13 of those finish under 6 grams of total fat for the entire shake — below the lower dose that produced a measurable effect in Brown and colleagues. Whatever absorption benefit exists in our lineup, we did not put it there on purpose. Our internal Score does not catch it either. Score averages 14.6 for recipes with no Functional slot, 15.2 for those with a botanical-only slot and 15.3 for those carrying a nut or seed, on an 11-to-20 scale — seven tenths of a point across a decision that moves total fat by nearly five grams. The Score does not track fat-soluble absorption and we should not imply that it does. The trials themselves are also small, seven and eleven subjects, run on salads rather than blended shakes, and neither used nuts as the lipid source. They establish that co-ingested fat matters at these doses. They do not establish what any particular MOMS recipe delivers.
What We Learned
The slot's name described our intent, not its contents. By weight the Functional slot is a nut-and-seed slot with a pinch of spice on top, and the fat we had been treating as incidental — as texture, as a few extra calories — has clearer human evidence behind it than the 0.5 grams of turmeric the slot is named for. A few grams of fat is the documented difference between negligible and measurable absorption of the carotenoids already sitting in the produce slot. The uncomfortable version is that the most defensible thing happening in our Functional slot is the thing we were not paying attention to, and that a name can keep you from looking at your own data for a long time.
What Changed at MOMS
We added a lipid-pairing check to recipe review. Any recipe whose Fruit and Vegetable or Body slot is carotenoid-dense — pumpkin, carrot, sweet potato, papaya, mango, spinach, kale — must clear 6 grams of total fat from whole-food sources before it ships, matching the lower effective dose in Brown and colleagues. That check flags 13 current recipes for reformulation. We are also renaming the column in our internal build sheet from "Functional" to "Fat and Functional", so that whoever is formulating sees what the slot actually does before reaching for 0.2 grams of a spice and counting it as a feature.
References
Brown MJ, Ferruzzi MG, Nguyen ML, Cooper DA, Eldridge AL, Schwartz SJ, White WS. Carotenoid bioavailability is higher from salads ingested with full-fat than with fat-reduced salad dressings as measured with electrochemical detection. American Journal of Clinical Nutrition, 80(2):396-403, 2004. | Unlu NZ, Bohn T, Clinton SK, Schwartz SJ. Carotenoid absorption from salad and salsa by humans is enhanced by the addition of avocado or avocado oil. Journal of Nutrition, 135(3):431-436, 2005.
Keep reading
Previous experiment
Next experiment
Not sure which shake fits you? We match the formulation to your body type.
Find your body type