Case study

We ran ERRA on our own brand first. It found errors we didn't know we had.

ERRA was built for moreU — the founder's own supplement brand. Before any customer touched it, we validated it against moreU's real formulas, printed labels, and supplier documents. The validation process caught three real errors in files we thought were clean.

The setup

moreU is a real, shipping supplement brand. Its formulation data lived where most brands' does: spreadsheets, supplier PDFs, a designer's label files, and email. Nothing looked broken. The labels were printed. The products were on shelves.

To validate ERRA, we entered moreU's formulas as structured data and compared the generated Supplement Facts panels line-for-line against the printed labels — and the source data against the supplier documents behind it.

What the process caught

Allergen gap #1
Before — in the files

An allergen present in a supplier's own documentation was missing from the working allergen declaration.

After — in ERRA

In ERRA, allergens live on the ingredient record and the declaration is derived from the formula — the gap surfaced the moment the data was structured.

Allergen gap #2
Before — in the files

A second allergen declaration didn't match what the supplier paperwork actually said.

After — in ERRA

Same mechanism: one canonical ingredient record, one derived declaration. There is no second place for the truth to drift.

Serving-count contradiction
Before — in the files

Two supplier documents for the same product disagreed about the serving count.

After — in ERRA

Entering both into one structured record forced the contradiction into the open — it had to be resolved before a panel could be generated.

None of these were caught by a person being more careful. They were caught by a system that derives declarations from structured data instead of trusting documents to agree with each other.

"Our files were fine" — so were ours

These errors sat in clean-looking, professionally produced documents for a brand whose founder cares deeply about getting this right. If your formulas live in spreadsheets and supplier PDFs, the honest assumption is that your files have a version of these too. The only way to know is to structure the data and let the contradictions surface.

Why this works

ERRA is the only cloud system where a Supplement Facts panel is derived from the approved, immutable formula version — validated against real printed labels — with a database-enforced audit trail an FDA auditor can't question.

One ingredient record carries the allergens, potency, and sourcing facts. The formula references it. The panel, allergen statement, and cost derive from the formula. When the inputs disagree, the system can't quietly render both versions — it forces a decision, and records who made it.

Run it on your files

We're onboarding five founding partner brands with white-glove data import — meaning we do this same validation exercise on your formulas and your printed labels, with you.

moreU is the founder's own brand — the only customer named here, because it's the only one whose story we can tell first-hand. ERRA performs deterministic regulatory math but does not provide regulatory or legal advice.