The American Nutrition Database is USDA’s FoodData Central, a free source for lab-based nutrient data and branded labels.
Basic
Better
Best
Match A Package
- Search by brand owner
- Use serving on the label
- Check publish date
Branded
Plan A Menu
- Pick survey-mapped items
- Use cups and pieces
- Mirror cafeteria portions
FNDDS
Build A Recipe
- Choose cooked form
- Apply yield factors
- Weigh in grams
Foundation
The national database in the U.S. blends lab science and grocery labels in one hub. You can look up a raw ingredient, a cooked dish, a store product, or a menu item and see nutrients in grams, milligrams, and IU when listed. The site is free, fast, and updated in waves. That mix makes it a solid starting point for home cooks, diet pros, trainers, and developers who build meal tools.
U.S. Nutrition Database Basics For Fast Results
The system groups data into a handful of sets. Each set exists for a reason. Picking the right one saves time and avoids mismatches.
| Data Set | What You Get | Best Use |
|---|---|---|
| Foundation Foods | Item-level lab data with sample notes, moisture loss, and yield factors | Precise cooking states and recipe checks |
| SR Legacy | Classic entries from the National Nutrient Database | Broad lookup when you want a fast baseline |
| FNDDS | Portion-based items mapped to What We Eat In America surveys | Meal plans, menus, and typical dishes |
| Branded | Packaged foods pulled from company records and market scans | Match a label by brand or UPC |
| Experimental/Survey | Research builds and specialty studies | Method notes and niche foods |
Start with a clear target. If you have a label in hand, use Branded search. If you’re cooking, pick Foundation for the cooked form and yield. If you’re logging meals for a group, FNDDS often feels closer to how people plate and pour.
Finding The Right Record Without Guesswork
Search by common name first, then filter by data type. Add a second term if needed—“raw,” “boiled,” “skinless,” or a brand. Open the record and scan four items: description, serving or gram weight, data type, and the publish date. Those four lines tell you if the entry matches what’s on your plate.
Serving sizes on packages use a federal set of reference amounts. These RACCs shape the serving line and dual-column labels on packages. When your task needs label alignment, stick to the serving and household units shown on the item.
Reading Nutrient Fields Like A Pro
Calories, macros, and fiber sit near the top in most records. Next come sugar and added sugar when listed, then fats by type, and sodium. Vitamins and minerals follow. Many Foundation entries include moisture and ash, which helps when you want dry-to-cooked comparisons. Branded items mirror the label, so some vitamins may be blank. That’s normal.
When an entry includes density or yield notes, you can convert a raw weight to a cooked portion with fewer surprises. That helps with grains, meats, and sauces that lose or gain water in the pan.
Label Rules That Shape The Numbers
Two anchors set the ground rules: the Nutrition Facts format and the serving size system. Both come from the FDA. If your work involves menu labels or a product launch, you’ll want to match those standards. The RACC list ties each food group to a base amount, and the label rules define which nutrients must appear and how rounding works. You’ll see the ripple effects in Branded entries inside the database.
Many shoppers scan for added sugars, sodium, and fiber first. The label highlights those lines for a reason. Developers who build apps around this data often surface the same lines to match how people read packages.
Accuracy Tips For Home Cooks And Builders
Pick the right form. Raw, cooked, drained, or with skin can swing calories and sodium. Use gram weights wherever possible. Household measures help, but grams remove guesswork when you scale recipes.
Match preparation. If you simmered beans with salt, avoid a “no salt added” entry. If you sautéed spinach, a raw entry will undercount shrinkage. Foundation entries often include yield factors that keep your math steady.
Watch publish dates. Branded items refresh when companies send updates. If an entry looks off, check for a newer record or pick a generic SR Legacy entry as a stand-in.
Power Features Many People Miss
Advanced filters save clicks. You can search by brand owner, ingredients, market channel, or date window. Food codes link FNDDS items to national survey dishes, which helps with menu mapping. Developers can also use the API to fetch records by FDC ID and return nutrient arrays in JSON for app use.
Foundation entries list sample locations and lot notes. That detail supports projects that need traceability or regional insight. It also explains small swings between entries that look the same at first glance.
From Pantry To Plate: A Simple Workflow
Here’s a clean way to run a meal project. Start with your ingredient list. For each item, pick a record that matches the form you’ll use in the pan. Pull gram weights. Cook, weigh the finished dish, and divide by portions. Compare your per-portion numbers to a matched FNDDS dish to sanity-check the result.
| Step | What To Check | Where To Look |
|---|---|---|
| Pick Form | Raw vs. cooked; drained; with or without skin | Description line |
| Confirm Weight | Grams per piece, cup, or slice | Measure section |
| Scan Notes | Yield, moisture, and lab details | Foundation footnotes |
| Align Serving | Match RACCs when you mirror labels | FDA serving rules |
| Log Version | Publish date and FDC ID for traceability | Header of record |
When To Use SR Legacy Versus Foundation
SR Legacy gives wide coverage with fewer sample notes. It’s handy when you want a quick baseline for a common item or a simple baked good. Foundation goes deeper. You’ll see moisture loss, brand and lot data, and sometimes cooking fat uptake. That detail helps with recipe math and claims work.
FNDDS For Real-World Portions
FNDDS entries tie to dishes people report in national surveys. The records carry household measures like cups and pieces. That makes them friendly for menu plans and cafeteria rollouts. If you want a value that mirrors how people plate food, start there.
Branded Records And When To Trust Them
Branded listings come from company submissions, market audits, and label images. They mirror packages. The strength is speed and brand match. The trade-off is limited micronutrients and less lab context. For general tracking, they’re fine. For research, pair them with Foundation or SR entries.
Common Pitfalls And Clean Fixes
Mixing raw and cooked states in one recipe skews totals. Fix it by converting all items to cooked or all to raw using yield factors. Swapping a salted canned item into a no-salt recipe inflates sodium. Fix it by draining and rinsing, then picking a matching record. Using cups for shredded cheese leads to wide spreads. Fix it by weighing.
How Developers Plug It Into Apps
The API returns foods, measures, and nutrients by FDC ID. You can filter by data type and pull only the fields you need. Start with a simple search endpoint, then store FDC IDs in your own tables for repeat calls. Respect rate limits and cache popular items. In user flows, surface serving sizes first, then the nutrient lines people use most: calories, protein, carbs, fiber, sugar, and sodium.
Why This Resource Still Matters
It blends lab science with lived eating patterns. That mix helps parents track meals, helps coaches set macro targets, and helps researchers audit recipes. It also gives small brands a path to align labels with federal rules without building a lab from scratch.
Quick Recap And Next Steps
Pick the right data set for the job. Read the description, gram weight, data type, and date before you log a single line. Use gram units, match cooked states, and lean on FNDDS for plate-ready portions. When a label match matters, apply FDA serving rules and scan added sugars. If you build tools, use the API and store FDC IDs for speed.
Learn more about FoodData Central data types. For label serving bases, see the FDA guidance on RACCs.