In this LAB note
Question · Hypothesis · Method · Results · What Failed · What We Learned · What Changed at MOMS
Question
Five grams of agave is the smallest deliberate addition we make to a shake. In a formula built to a fixed calorie target, does putting a sweetener in the Flavor slot add sugar to the shake - or does it change what else has to be in it?
Hypothesis
We expected the Flavor slot to be a small, honest cost. A sweetener is sugar, sugar is carbohydrate, and adding one should move the shake a gram or two in the wrong direction in exchange for making it drinkable. We assumed the unsweetened recipes would be the leaner ones, and that the case for the Flavor slot would have to be made on palatability rather than nutrition. We expected to be defending a compromise. The database says we had the sign backwards.
Method
Rogers and colleagues (International Journal of Obesity, 2016) found that when low-energy sweeteners replace sugar rather than add to it, both energy intake and body weight fall. Alkaabi and colleagues (Nutrition Journal, 2011) found five date varieties to be low-GI, while dates remain about 75% carbohydrate by mass. Internally, we parsed the Flavor column of all 201 recipes into caloric sweetener, monk fruit, aromatic-only, and empty-slot groups, then compared macros, Score, date usage, and matched recipes.
Results
The Flavor slot is filled in 153 of 201 recipes. Agave leads at 66 uses, then monk fruit at 51, maple syrup at 27, lemon at 16, honey at 12 and cinnamon at 12. The median caloric sweetener dose is 5g. That 5g is the entire intervention, and here is what sits on either side of it. The 48 recipes with an empty Flavor slot average 376.40 kcal, 33.21g protein, 43.92g carbohydrate, 8.48g fat, 7.84g fiber, Score 14.27. The 105 caloric-sweetener recipes average 381.58 kcal, 32.98g protein, 39.38g carbohydrate, 10.92g fat, 7.73g fiber, Score 15.20. The 51 monk fruit recipes average 358.96 kcal, 33.11g protein, 34.36g carbohydrate, 10.55g fat, 7.36g fiber, Score 14.82. The unsweetened recipes are the highest-carbohydrate group in the database. Adding a caloric sweetener is associated with 4.54g less carbohydrate; adding monk fruit with 9.56g less, and 17.4 fewer calories. Protein sits within 0.23g across all three groups. The mechanism is in the date column. Among recipes with an empty Flavor slot, 85.4% contain dates, at a median dose of 15g and a mean of 14.87g across the whole group. Among caloric-sweetener recipes, 31.4% contain dates, mean dose 3.43g. Among monk fruit recipes, 19.6% contain dates, mean 3.10g. A shake with nothing in the Flavor slot is not an unsweetened shake. It is a shake sweetened with roughly 15 grams of dates, which is a nutritionally respectable decision and an expensive one in carbohydrate - about 11g of it. Five grams of agave does the same sensory job for roughly 3.5g. A matched pair makes it concrete. Sunrise Surge (EY-W1-ENERGY-001-V1) and Bright Start (EY-W1-ENERGY-005-V1) share body type, week, goal and the Filling label, and land 3.1 kcal apart. Sunrise Surge carries Maple Syrup 5g, no dates, and 195ml of coconut milk over banana, beet and carrot: 393.5 kcal, 37.9g carbohydrate, 34.5g protein, 11.7g fat, Score 16. Bright Start has an empty Flavor slot and reaches for fruit instead - Dates 10g, Kiwi 44.4g, Asian Pear 85g, Cherry 42.5g on coconut water: 390.4 kcal, 55.6g carbohydrate, 33.4g protein, 4.8g fat, Score 14. Same calories, 17.7 grams of carbohydrate apart. The unsweetened one is the sweeter one.
What Failed
The dose-response, and the assumption that the effect was about sweetener at all. If a sweetener displaces carbohydrate, more sweetener should displace more. Within the 105 caloric-sweetener recipes, sweetener dose correlates with total carbohydrate at r = +0.24 - the wrong sign, and weak. The effect is categorical, not proportional: what matters is whether the formula reached for a concentrated sweetener, not how much it used. Worse, we found 33 recipes that took the sweetener and kept the dates. They average 42.41g carbohydrate against 38.00g for the 72 sweetened recipes that dropped dates entirely - almost all the way back to the 43.92g unsweetened baseline. In those 33 the Flavor slot was added on top of the existing sweetness rather than in place of it, which is precisely the failure mode Rogers' review warns about: substitution works, addition does not. And the clearest counterexample is a recipe where the sweetener governs nothing at all. Rooted Calm (MY-W3-GUT-004-V1) carries Agave 10g, twice the median dose, no dates, and is the highest-carbohydrate recipe in the entire sweetened set at 62.0g - 18 grams above its own group's average, on 396.5 kcal, Score 13. The carbohydrate arrived through fruit and liquid, and no amount of agave in the Flavor slot was going to reach it. The sweetener sets what the formula no longer has to do. It does not set the shake.
What We Learned
The Flavor slot is not an indulgence. Sweetness has to come from somewhere, and whole-food sweetness arrives bundled with carbohydrate. A concentrated sweetener can buy the same sensory result for a quarter of the mass only when it replaces something. Dates remain a good ingredient with low-GI credentials, fiber, and potassium; the finding is that no added sweetener was never a description of a leaner shake.
What Changed at MOMS
Every recipe now declares its sweetness source explicitly: concentrated, whole-food, or none. Adding a sweetener requires removing the whole-food sweetness it replaces in the same edit. Monk fruit is the default Flavor ingredient for Light and Detox goals, while dates remain deliberate in Build and Energy recipes where fiber and slower arrival are worth the carbohydrate. The both-ingredient recipes are queued for reformulation.
References
1. Rogers PJ, Hogenkamp PS, de Graaf C, Higgs S, Lluch A, Ness AR, Penfold C, Perry R, Putz P, Yeomans MR, Mela DJ. Does low-energy sweetener consumption affect energy intake and body weight? International Journal of Obesity. 2016;40(3):381-394. 2. Alkaabi JM, Al-Dabbagh B, Ahmad S, Saadi HF, Gariballa S, Ghazali MA. Glycemic indices of five varieties of dates in healthy and diabetic subjects. Nutrition Journal. 2011;10:59.
Keep reading
Previous experiment
Not sure which shake fits you? We match the formulation to your body type.
Find your body type