A food database stores nutrition entries for quick search and logging. Database quality sets the ceiling on how accurate your calorie and macro totals can be. For the full audit workflow on catching bad entries, raw-versus-cooked mismatches, and restaurant underestimation, read Food Database Accuracy: Why Your Macro Numbers Drift and How to Audit Them. For photo-based workflows, see Food Logging.
01Entries and accuracy
| Entry type | Source | Tip |
|---|---|---|
| Packaged foods | Label data | Match serving size and brand |
| Restaurant items | Brand or chain data | Portions vary; sanity check calories |
| Generic foods | Standard references | Use for simple single-ingredient items |
| Custom recipes | User-built from ingredients | Log once, reuse often |
02User correction workflow
When values look wrong, use a controlled correction step.
| Error type | User action |
|---|---|
| Missing serving data | add serving context before saving |
| Suspicious calorie mismatch | compare alternate entries and annotate context |
| Portion ambiguity | save a custom corrected version |
03When to override automatically
Manual overrides are most useful when:
- Serving sizes don’t match across brands or entries.
- Label data is missing or inconsistent, creating repeated logging friction.
- You’re in a tightly constrained meal planning phase and need repeatable numbers.
Prefer a corrected custom food saved once over repeated guesswork when source quality is uncertain.
If your weekly trend does not match your logged intake, pair this page with Calorie Counting Accuracy and the full Food Database Accuracy guide.
