Nutritional Facts Panel Testing: What Third-Party Labs Actually Measure — and Where Labels Most Often Fail

Every year, the FDA issues warning letters to food and supplement companies for nutritional labeling violations. Some of those violations involve companies that genuinely didn’t know their labels were wrong — they commissioned a contract manufacturer, received a finished product and a spec sheet, and printed a Nutrition Facts panel they thought was accurate. The lab data told a different story.

Third-party nutritional facts panel testing exists precisely to catch that gap. But a lot of brands don’t fully understand what labs are actually measuring, why certain nutrients are harder to verify than others, or where their existing data could still leave them exposed. What follows is a practical look at what a full label verification actually involves — from FDA’s underlying requirements through the nutrients that consistently generate surprises.

What FDA’s 21 CFR Part 101 Actually Requires (and What It Doesn’t Say)

The foundational rule is deceptively simple: your Nutrition Facts panel must accurately represent the product you’re selling. Under 21 CFR § 101.9, nutrients must be determined by appropriate analytical methods — not pulled from theoretical formulation values, and not taken directly from ingredient supplier certificates of analysis.

What surprises many companies is that FDA doesn’t impose a blanket ±20% tolerance for all nutrients. The compliance picture is considerably more nuanced:

For macronutrients and calories, FDA applies a reasonableness standard. A product labeled at 200 calories that tests at 210 is unlikely to trigger an enforcement action. A product consistently testing at 280 calories is a different matter — one that brands have litigated over.

For vitamins and minerals, the 2016 Nutrition Facts label update — which large manufacturers had to comply with by January 1, 2020, and companies with annual food sales under $10 million by January 1, 2021 — changed which nutrients are mandatory. Vitamin D and potassium are now required. Vitamins A and C were removed from the mandatory list. FDA guidance permits manufacturers to add overage at the time of production to account for degradation over shelf life, but the labeled value still can’t exceed actual content by a substantial margin at the end of shelf life.

For dietary supplements governed by DSHEA, 21 CFR § 101.36 creates a parallel requirement: the Supplement Facts panel must reflect the amount of each ingredient as declared. FDA has consistently held that supplement manufacturers must verify label claims through appropriate testing — not supplier documentation alone.

Products manufactured before the 2020/2021 compliance deadlines that still carry the old Nutrition Facts format are out of compliance. We still see these in the market, particularly from smaller brands that launched pre-2020 and haven’t updated their packaging.

What a Full Nutritional Testing Battery Actually Includes

When a brand submits a product to a food safety testing laboratory for Nutrition Facts panel verification, the typical battery runs 20 to 35 individual analyte determinations, depending on what’s declared on the label. These fall into several analytical categories.

Proximate analysis is the foundation. Moisture content is measured by oven drying at a defined temperature, and all downstream calculations are expressed on a wet-weight basis — so moisture error compounds into every other result. Ash content comes from high-temperature incineration and feeds into mineral recovery calculations. Total fat is quantified by acid hydrolysis followed by gravimetric measurement or GC/FID — one of the more method-sensitive determinations, particularly for products with emulsified or matrix-bound fats. Protein is measured using Kjeldahl or Dumas combustion; Kjeldahl quantifies total nitrogen and applies a conversion factor (typically 6.25 for most food matrices) to arrive at protein content.

Carbohydrates and fiber introduce the most analytical complexity. Total carbohydrate is generally calculated by difference — 100g minus moisture, fat, protein, and ash — which means errors in the other analytes compound directly into the carbohydrate number. Dietary fiber is measured using AOAC 985.29 or the more comprehensive AOAC 2009.01 enzymatic-gravimetric method. This is where we see the most significant disagreements between formulation records and actual test results, particularly in products containing prebiotic fibers like inulin, FOS, or resistant starch. These fiber types register differently depending on which method a lab uses, and brands frequently discover their supplier CoA and their third-party test report reflect numbers from different methods.

Added sugars — introduced as a required label element by the 2016 update — present their own challenge. In some products, analytical chemistry can’t distinguish added sugar from naturally occurring sugar. When that’s the case, the manufacturer’s records become part of the compliance picture alongside test data.

Sodium is measured by ICP-OES and is one of the most reliable analytes in a nutritional panel. It’s stable, well-characterized, and closely tracks formulation intent. It’s also a nutrient FDA pays close attention to in processed foods.

Fat-soluble vitamins (A, D, E, K) are measured by HPLC with UV or fluorescence detection. Vitamin D deserves particular attention: it’s susceptible to degradation in gummy and softgel formats, and we consistently see end-of-shelf-life values 15–30% below label claims in products that didn’t account for this in the formulation overage strategy. Water-soluble B vitamins (B1, B2, B3, B6, B12, folate, pantothenic acid, biotin) are also HPLC-based, though folate and B12 at typical dietary supplement concentrations require more sensitive detection methods.

Minerals — calcium, iron, potassium, magnesium, zinc, phosphorus, and others — are measured by ICP-OES in a single analytical run. Results here tend to be reproducible and track closely to formulation expectations, unless there’s a raw material substitution or supplier adulteration issue upstream. Potassium is worth noting given its new mandatory status; many brands labeled for years without quantifying it and are discovering they need to now.

Where Nutrition Labels Fail Most Often

After running thousands of label verification panels across food and supplement products, some failure patterns are consistent.

Dietary fiber is the most common discrepancy. Brands formulating with novel fiber ingredients — chicory root extract, psyllium husk, high-amylose corn starch — often rely on supplier CoAs generated with one validated method. Their testing lab uses a different validated method. The results don’t agree. This isn’t fraud; it’s method variation. But it still produces a panel that doesn’t reflect the tested product, which is what FDA looks for.

Vitamin D is the most common degradation-related failure. Gummy formats are particularly vulnerable. The combination of processing heat, elevated water activity, and ambient storage time produces measurable potency loss. Brands that formulate without an overage strategy — typically 10–25% above label depending on format and target shelf life — will find their products testing below label by month 12 to 18. We routinely see gummies labeled at 2,000 IU per serving testing at 1,400–1,600 IU after six months of ambient storage conditions.

Added sugars are frequently miscalculated. Many brands — especially those that launched products before 2020 and have since updated their labels — are still calculating added sugars from recipe records rather than verified analytical data. FDA can and does flag this discrepancy during routine inspections.

Protein content from Kjeldahl can be overstated when a product contains non-protein nitrogen sources: free amino acids added in crystalline form, creatine monohydrate, or collagen (which has an atypical amino acid composition that makes the standard 6.25 conversion factor inaccurate). Brands relying on Kjeldahl results for these product types may be labeling 5–8% more protein than the actual dietary protein equivalent.

Calorie calculations in high-fat specialty products — ketogenic formulas, full-fat dairy concentrates, nut-based energy bars — can drift significantly when fat measurement has even minor analytical variance. A 5% relative error in fat content for a product with 40g fat per 100g translates to approximately 18 additional calories per serving. Across an entire product line, this adds up to a systematic labeling problem.

Getting More From Your Testing Data

A few observations from the lab side that aren’t obvious until you’ve seen a lot of results:

Submit finished product, not raw material CoAs. We receive quote requests from brands wanting to verify their Nutrition Facts panel using ingredient-level CoAs from their contract manufacturers. CoAs tell you what went into the formula. They don’t tell you what came out of it. Moisture loss during spray drying, vitamin degradation during high-temperature processing, fat fractionation during emulsification — all of these change the final nutrient profile. Always test the finished product in its final packaging format.

Test at multiple time points if you’re making a shelf life claim. The snapshot you get at day zero is not your worst-case compliance scenario. FDA scrutiny of food and supplement labels often focuses on market-ready or end-of-shelf-life products. If your vitamin D is tight at manufacture, it may not be compliant by month 12. Build a stability-informed testing schedule rather than a one-time confirmation test.

Understand method sensitivity before comparing results across labs. If your contract manufacturer’s CoA used AOAC 2001.02 for total dietary fiber and your ISO 17025-accredited testing lab used AOAC 985.29, you will see different numbers from the same product. This doesn’t indicate error by either party — it reflects the genuine method-dependence of fiber measurement. An accredited lab should be able to tell you exactly which method generated each result and what the expected interlaboratory precision is for that method.

Use the data to drive formulation decisions, not just to pass a single test. The purpose of label verification isn’t to confirm accuracy once and file the CoA. It’s to understand where your product is analytically vulnerable — which nutrients degrade, which manufacturing variables introduce potency variance — and build a testing schedule around that knowledge.

The brands that avoid costly reformulations and FDA Warning Letters aren’t the ones with the most elaborate labels. They’re the ones testing early enough in the product lifecycle to fix problems before products hit shelves.

---

Written by Nour Abochama, Vice President of Operations, Qalitex Laboratories. Learn more about our team

Talk to our team about your testing needs. Contact us

Related from our network

• FDA Compliance Strategy for Food and Supplement Manufacturers — Regulatory consulting on labeling requirements, 21 CFR compliance, and FDA inspection readiness
• Raw Material and Ingredient CoA Verification Testing — Third-party verification of supplier ingredient data for global B2B manufacturers