When Compliance Becomes a Formulation Problem

For much of the last thirty years, regulatory work inside a large food manufacturer followed a familiar pattern.

R&D developed the product. Regulatory affairs reviewed the finished specification, prepared the required documentation, and answered questions from customers or auditors. The handoff between the two functions was usually a document.

That model is under strain.

The issue is not simply that food regulation has become stricter. It is that the answer to a seemingly simple question, “Can we use this ingredient here?”, now changes more often, varies more by jurisdiction, and can arrive with a reformulation deadline attached.

That makes compliance an R&D constraint. It has to be understood while the product is being designed, not after the formula is finished.

‍

Why Are Food Additive Rules Creating R&D Work?

A regulatory change can now turn a long-established formulation into an active development project.

In January 2025, the FDA revoked the authorization for FD&C Red No. 3 in food and ingested drugs. Food manufacturers have until January 15, 2027 to reformulate affected products, while manufacturers of ingested drugs have until January 18, 2028. FDA’s guidance on FD&C Red No. 3 makes the practical implication clear: an ingredient once used in a finished product now comes with a hard product-development deadline.

The same year, the FDA and the Department of Health and Human Services announced measures to work with industry to phase out petroleum-based synthetic dyes from the US food supply. The details will continue to evolve, but the direction of travel matters to food and beverage teams. Ingredients that have been stable parts of a formulation can become candidates for replacement, testing, and relabeling with very little warning. FDA’s color-additive information tracks the broader initiative.

A withdrawn authorization is a regulatory event. Replacing the ingredient is a formulation project.

‍

Why Does One Country No Longer Mean One Rule?

The regulatory map is also becoming more complex inside individual countries.

California’s Food Safety Act prohibits the sale of foods containing Red Dye No. 3, propylparaben, brominated vegetable oil, or potassium bromate from January 1, 2027. The California Food Safety Act was initially a state-level story. It did not stay one.

In 2025, more than 140 food-additive bills were introduced across 38 US states. Eleven school-food restriction bills were enacted across Arizona, Delaware, Louisiana, Tennessee, Texas, Utah, Virginia, and West Virginia. West Virginia was the first state to put restrictions into effect covering both school and retail settings. MultiState’s 2025 review shows how quickly the issue has moved beyond one jurisdiction or one additive.

The substances on these lists overlap, but they are not identical. The effective dates differ. The scope differs. Some restrictions apply to school food. Others reach retail products. A portfolio team cannot treat “compliant in the US” as a single answer if the products move across state lines.

That is before international markets enter the picture.

Titanium dioxide is a useful example. The European Union withdrew authorization for titanium dioxide, also known as E171, as a food additive in 2022. In the United States, FDA regulations continue to allow its use as a color additive in foods, subject to specified conditions. The European Commission’s notice and the FDA’s titanium dioxide guidance describe two very different regulatory positions on the same substance.

For a formulator supplying both markets, that is not an abstract regulatory difference. It is a different formulation decision.

‍

Why Does a Regulatory Change Trigger So Much Testing?

Ingredients such as colors, preservatives, and stabilizers are rarely interchangeable.

Replacing a synthetic color with a plant-based alternative can change heat stability, pH sensitivity, appearance, and shelf life. Removing or changing a preservative can affect the microbial model, process conditions, validation requirements, and final product specification.

The substitution itself may be straightforward. Proving that the new formula still meets sensory, stability, safety, quality, and production requirements is not.

That work can mean new bench trials, stability studies, sensory panels, shelf-life testing, process adjustments, packaging checks, specification updates, and market-specific documentation. Then the work has to be repeated for every product affected by the same ingredient change.

This is where the portfolio question becomes more important than the individual formula.

When an ingredient is restricted, the first question is not “What can replace it?” The first question is “Which products contain it, in which markets, at what level, and under which rules are they affected?”

For many organizations, that first question takes longer than it should. Ingredient information may live in one system, formulations in another, specifications in static documents, and market requirements in a regulatory team’s working files.

The regulatory deadline does not wait for that reconstruction.

‍

Why Is Ingredient Compliance Harder Than a Lookup?

The intuitive response is to create a list of restricted substances by country and circulate it.

Lists are useful. They are not enough.

A compliance limit is rarely a simple property of an ingredient. It is usually a relationship between an ingredient, its concentration, the food category, the jurisdiction, and the conditions of use.

The Codex General Standard for Food Additives is structured around this reality. It uses a food-category system and sets maximum use levels for specific additives in specific food groups. The same additive can have different conditions of use in a beverage, a confectionery product, or a dairy dessert.

That means the question is not simply, “Is this ingredient permitted?” It is closer to, “Is this ingredient permitted at this level, in this type of product, for this market, under this specific set of conditions?”

The calculation basis matters as well. Codex generally expresses maximum use levels on the final product as consumed, unless otherwise specified. Carryover from a compound ingredient can create a separate question again. Two people can start with the same formulation and reach different answers if they are applying different assumptions about the basis of the limit.

The source material itself creates another layer of complexity. The authoritative version of a requirement may sit in a national register, in a local language, with exemptions and scope clauses that are more important than the headline rule. A quick English summary may be useful for orientation, but it cannot replace the source when the decision is high stakes.

Then there is interpretation.

Regulatory affairs teams build institutional knowledge over time. They know where a regulation is clear, where it is ambiguous, and where their organization has an established position based on prior reviews or conversations with regulators. That judgment should not be buried in email or overridden by a generic database entry. It should be explicit, attributable, and available to the teams making formulation decisions.

‍

What Does a Better Compliance Workflow Look Like?

The organizations handling this well have changed where the compliance question sits.

They treat regulatory status as structured data, rather than as a PDF or a spreadsheet stored outside the development process. That makes it possible to connect an ingredient to the relevant jurisdiction, food category, limit, source, and interpretation.

They can identify affected products quickly. When a substance changes status, the team can see which products use it, where those products are sold, and which formulations need attention before the work begins.

They keep the evidence close to the answer. A compliance flag should lead back to the source provision and the internal interpretation behind it. If regulatory affairs cannot see where the answer came from, the flag will not survive review.

Most importantly, they bring the constraint into formulation early enough to change the direction of work. The least expensive time to find out that an ingredient choice will not work in a target market is before the stability trial, not after the final formula has been prepared.

This is also why connected formulation records matter. When ingredient composition, process, performance, and product context stay together, teams can evaluate a regulatory change as part of the development decision rather than rebuild the context from disconnected files. Food and beverage formulation PLM is one way teams can keep those dependencies visible.

‍

The Shift Is About Timing

This is not only a story about food additives.

It is a story about the tempo of regulation.

When rules changed slowly, it was efficient to hold regulatory knowledge in a specialist function and consult that function near the end of development. When authorizations reopen, state rules diverge, and the same ingredient has different status in two target markets, compliance needs to shape the work from the beginning.

The companies that adapt fastest will not necessarily be the ones with the largest regulatory teams. They will be the ones where formulation, regulatory, and product data can be connected quickly enough to turn a changing requirement into a clear development decision.

For a broader look at how connected formulation, process, and quality data supports food and beverage R&D, see the F&B R&D Data Playbook.