Skip to content
Analyst work

Data study · October 2026

The protein boom

Question

Does the rush of high-protein products match a real gap in what people eat?

Headline finding

On average, no. Average intake in every age and sex group runs 1.25 to 1.74 times the recommended amount (the RDA). The shortfall sits with girls 14 to 18, where 20% fall below the requirement, and with adults 71 and older at 12% of men and 11% of women. Gen Z adults spend 2.8 times what Boomers spend on protein products.

My recommendation

Aim the new capacity at adults 65 and older and the family members who shop for them. Core Power keeps its athlete advertising at today’s budget.

Tools
  • CSV tables with a source URL on every row
  • JavaScript on Node for the derived rates
  • D3.js for the first dashboard
  • React and SVG for the charts on this page

The question

The client

fairlife, the dairy brand owned by The Coca-Cola Company. Coca-Cola paid $0.98 billion for full ownership in 2020 and a final milestone payment of $6.2 billion in 2025. A $0.65 billion plant in Webster, New York opens in 2026, and Coca-Cola announced a $0.65 billion expansion in Coopersville, Michigan in March 2026.

The decision they face

Which buyers should fairlife aim that new capacity at?

The data, 2 tables from 24 sources

  • figures.csv181 rows

    Every sourced figure in the study, one per row. It covers mean protein intake by age and sex, usual-intake percentiles, the share of each group below the requirement, the RDA and EAR, the 2026 guideline range, survey answers, market tracker figures and Coca-Cola’s records on fairlife.

    Where it came from
    USDA’s What We Eat in America and NHANES usual-intake tables, the National Academies and Health Canada reference tables, surveys from IFIC, Empower, The New Consumer and Circana, trackers from Innova, Mintel, Beverage Digest and YPulse, and Coca-Cola filings and press coverage. That is 24 named sources at 22 URLs.
    Why I chose it
    Every figure shares one ID scheme, so the code can divide an intake row by a yardstick row.
    What it cannot tell us
    Line up in time. The means cover 2021 to 2023 and the shares below the requirement cover 2015 to 2018. Several market figures reach the file through trade press, and 9 rows are Coca-Cola or fairlife reporting on itself.
  • drivers.csv12 rows

    12 drivers behind the options. Each has what the data says, the option it favors, a strength score from 1 to 3 and the IDs of the rows it rests on.

    Where it came from
    My own coding. Strength runs from 1 for a single projection to 3 for official data or a direct company record.
    Why I chose it
    It ties each option to rows a reader can open.
    What it cannot tell us
    Stand as measurement. The scores and the option each driver points to are my judgment, and the evidence IDs are there so a reader can disagree.

The 24 sources behind figures.csv

figures.csv holds figures from 24 sources on 22 pages. Each one is listed with the rows it supplies and what they cover. 5 of them carry figures a company reported about itself.

Show the 24 sources

The process

  1. One row per figure

    Each figure went into a row of its own with an ID, a source name, a URL, the date I retrieved it and a flag for whether the company reported it about itself. That came to 181 rows. I read the USDA figures from published PDFs.

  2. Match the age bands

    What We Eat in America reports ages 12 to 19 as one band, and the RDA is set for 14 to 18. I set the 12 to 19 mean against the 14 to 18 RDA. Every adult band uses the RDA for 19 and older, 56 g a day for men and 46 g for women.

  3. Interpolate

    USDA publishes seven percentiles of usual intake per kilogram for each group. The 2026 Dietary Guidelines set a range that starts at 1.2 g/kg, and no table reports the share below it. I drew a straight line between the two percentiles on either side of 1.2 and read the share off it.

  4. Join

    The spend survey reports by generation and NHANES reports by age band. I matched Gen Z to 19 to 30, Millennials to 31 to 50, Gen X to 51 to 70 and Boomers to the average of 51 to 70 and 71 and older. Men and women count equally in each band. Gen X and Boomers share a band, so the match is loose.

  5. Check

    I ran the same interpolation at each group’s EAR and compared it with the share USDA publishes. 10 of 12 groups could be checked, and each landed within 0.8 points. The other 2 have an EAR below their 5th percentile. The check covers the low end of the curve. The 1.2 g/kg line falls between the 25th and 75th percentiles, where the published points sit further apart, so those shares can be off by more.

Boomers before and after the join.

Before, 5 rows of figures.csv (6 of 12 columns shown)
metric_id,metric,sex,age_group,value,unit
S_EMP_SPEND_BOOM,Weekly spend on protein products,All,Boomers,27,USD/week
UE_M_5170,Share with usual intake below EAR,Male,51-70,5,%
UE_F_5170,Share with usual intake below EAR,Female,51-70,8,%
UE_M_71P,Share with usual intake below EAR,Male,71+,12,%
UE_F_71P,Share with usual intake below EAR,Female,71+,11,%
After, one row per generation
{
  "generation": "Boomers",
  "bands": "51-70 and 71+",
  "spend": 27,
  "pct_below_ear": 9
}

Boomers have no NHANES band of their own. I averaged men and women in each band, then averaged the two bands. USDA publishes these shares as whole numbers, so the result is coarse.

The code for the key steps

The interpolation behind every share under the 1.2 g/kg floor. derive.mjs
export function shareBelow(cut, xs) {
  let pct = cut <= xs[0] ? P[0] : P[6];
  for (let i = 0; i < 6; i++) if (cut > xs[i] && cut <= xs[i + 1]) pct = P[i] + ((P[i + 1] - P[i]) * (cut - xs[i])) / (xs[i + 1] - xs[i]);
  return pct;
}
The join between survey generations and NHANES age bands. derive.mjs
export function genAlign(figures) {
  const m = pick(figures);
  const avg = (a, b) => (a + b) / 2;
  const band = (k) => avg(m(`UE_M_${k}`), m(`UE_F_${k}`));
  return [
    { generation: "Gen Z", bands: "19-30", pct_below_ear: band("1930"), spend: m("S_EMP_SPEND_GENZ") },
    { generation: "Millennials", bands: "31-50", pct_below_ear: band("3150"), spend: m("S_EMP_SPEND_MILL") },
    { generation: "Gen X", bands: "51-70", pct_below_ear: band("5170"), spend: m("S_EMP_SPEND_GENX") },
    { generation: "Boomers", bands: "51-70 and 71+", pct_below_ear: avg(band("5170"), band("71P")), spend: m("S_EMP_SPEND_BOOM") },
  ];
}

Dead ends

  • Google Trends

    Press coverage of The New Consumer’s report says US searches for protein hit an all-time high in March 2026. I could not pull the raw Google Trends series, so the study holds that claim as one second-hand row and draws no search chart.

  • Newer intake tails

    I wanted the low tail for the same years as the means. The latest usual-intake tables by sex and age that I could retrieve cover 2015 to 2018, so the tails and the 2021 to 2023 means come from different survey rounds.

The 3 findings that change the decision

  1. Finding 1

    1.25x

    Lowest average intake against the RDA, females 12 to 19. The highest group sits at 1.74x

    Average protein intake in 2021 to 2023 sat between 1.25 and 1.74 times the recommended amount (the RDA) in all 14 age and sex groups, so no group falls short on average.

    Dataset figures.csv
    Query derive.mjs, intakeRatio()

    Every group’s average intake clears its recommended amount, from 1.25 to 1.74 times

    Mean grams a day from What We Eat in America, August 2021 to August 2023, divided by the RDA in grams a day. A value of 1 means the mean equals the RDA. The 12 to 19 band is set against the 14 to 18 RDA. A mean above the RDA can still hide a low tail, which the next chart shows.

    Mean intake as a multiple of the RDA

    Source: figures.csv, derive.mjs, intakeRatio().

    Show the numbers as a table
    GroupMean intake as a multiple of the RDAMean intake (g a day)RDA (g a day)
    Females 12 to 191.2557.546
    Females 60 to 691.3260.546
    Females 70 and older1.3260.746
    Males 70 and older1.3877.356
    Females 50 to 591.4064.646
    Females 30 to 391.4566.846
    Females 40 to 491.4868.346
    Males 60 to 691.5385.456
    Females 20 to 291.5872.746
    Males 50 to 591.5989.156
    Males 30 to 391.5989.356
    Males 12 to 191.6183.852
    Males 20 to 291.6994.856
    Males 40 to 491.7497.656
  2. Finding 2

    20%

    Girls 14 to 18 below the requirement, against 8% of adult women

    20% of girls 14 to 18 usually eat less protein than they need, 2.5 times the rate for adult women. Adults 71 and older come next, at 12% of men and 11% of women.

    Dataset figures.csv
    Query derive.mjs, earRows()

    20% of girls 14 to 18 fall below the protein requirement

    Share of each group whose usual intake per kilogram of body weight falls below the Estimated Average Requirement (EAR), the level that covers half of a group’s needs. NHANES 2015 to 2018. The teen and 71 and older estimates carry the widest standard errors in USDA’s table.

    Share below the EAR (%)

    Source: figures.csv, derive.mjs, earRows().

    Show the numbers as a table
    GroupShare below the EAR (%)EAR (g/kg a day)Median usual intake (g/kg a day)
    Females 14 to 1820%0.711.03
    Males 71 and older12%0.661.07
    Females 71 and older11%0.661.01
    Males 14 to 189%0.731.29
    Females 51 to 708%0.661.08
    Females 31 to 507%0.661.10
    Females 19 to 306%0.661.12
    Males 51 to 705%0.661.24
    Males 19 to 304%0.661.29
    Males 31 to 504%0.661.26
  3. Finding 3

    2.8x

    Weekly protein spend, Gen Z at $75 over Boomers at $27

    Gen Z adults report $75 a week on protein products and Boomers report $27, yet Boomers are the more likely to fall short, 9.0% against 5.0%.

    Dataset figures.csv
    Query derive.mjs, genAlign()

    Gen Z spends 2.8x what Boomers spend on protein products

    Weekly spend comes from one online survey of adults, Empower’s in September 2025. Hover, tap or focus a bar for the share below the EAR in the matched NHANES age bands, 2015 to 2018, with men and women averaged.

    Weekly spend on protein products (US dollars)

    Source: figures.csv, derive.mjs, genAlign().

    Show the numbers as a table
    GenerationWeekly spend on protein products (US dollars)Share below the EARNHANES age bands
    Gen Z755.0%19-30
    Millennials675.5%31-50
    Gen X486.5%51-70
    Boomers279.0%51-70 and 71+

My recommendation

Aim the new plants at adults 65 and older and the family who shop for them.

  1. A

    Healthy-aging line

    Position Nutrition Plan and fairlife milk for adults 65 and older and the family members who shop for them.

    Evidence
    Finding 2. Adults 71 and older are the second most likely group to fall short, at 12% of men and 11% of women. An estimated 71% of women 71 and older eat less than 1.2 grams per kilo of body weight, the low end of the 2026 Dietary Guidelines, and 33% of adults link protein to healthy aging.
    Main risk
    Boomers spend $27 a week on protein products today, the least of any generation in finding 3, so fairlife would have to build this demand.
  • Older adults have a real protein gap. Adults 71 and older are the second most likely group to fall short of the requirement, and most women that age eat less than the low end of the 2026 Dietary Guidelines.
  • The risk is demand. Boomers spend the least on protein products today, so the ads have to reach the adult children who shop for them too.

Limits and next steps

What the data cannot support

  • The answer depends on the yardstick. Against the EAR, 5% of adult men and 8% of adult women fall short. Against the 1.2 g/kg low end of the 2026 Dietary Guidelines my interpolation puts it near 46% and 62%, and each estimate can be off by a few points.
  • The shares below the requirement come from 2015 to 2018. The share of adults trying to eat protein rose from 62% in 2021 to 70% in 2025, after that data was collected, so I cannot say whether the gap has closed.
  • Nothing here links one person’s spending to that person’s intake. The spend survey covers adults only, and the group with the largest gap is teenage girls.
  • The driver scores are my judgment. 9 of the 181 figures are Coca-Cola or fairlife reporting on itself, and several market figures reach the file through trade press.

What I would test with more access

  • With the NHANES 2021 to 2023 microdata I would rebuild the usual-intake tails and see whether girls 14 to 18 and adults 71 and older still lead.
  • I would pull the raw Google Trends series and set it beside the launch counts.
  • With fairlife’s sales data I would look at who buys Nutrition Plan today by age of household, which option A turns on.

TL;DR

  1. Studied

    Whether the rush of high-protein products matches a real gap in what people eat.

  2. Found

    On average, no. Average intake runs 1.25 to 1.74 times the recommended amount in every age and sex group.

  3. By the numbers

    Lowest group’s average intake against the recommended amount
    1.25x
    Girls 14 to 18 below the requirement
    20%
    Gen Z protein spend over Boomers
    2.8x
  4. The catch

    The answer turns on the yardstick. By my estimate, near 46% of adult men and 62% of women fall short of 1.2 grams per kilo.

  5. Next

    With person-level NHANES 2021 to 2023 data, re-estimate who falls short and check which groups still lead.