Better evidence, shared responsibly.

Food manufacturers know their recipes, pack formats, production limits, labelling rules and serving-size decisions. PerBiscuit can add a different kind of evidence: anonymous, aggregated information about the portions people actually report eating or drinking.

Looking at those two sources together can show where serving guidance is working, where it is being misunderstood, and where the label may no longer match how the product is commonly used.

The aim is not to attack manufacturers. The aim is to improve the quality of portion information available to everyone.

What PerBiscuit can add

PerBiscuit can help build a clearer picture of the gap between label servings and real-world portions.

That might include:

• How reported portions compare with the serving shown on a label
• Which product categories have the largest portion gaps
• Which pack formats make portions easier or harder to understand
• Where on-pack wording, diagrams or measuring guidance could be clearer
• Which products already give realistic and useful portion guidance
• How similar products compare within the same category
• Where better consumer information could support better product design

This is not about calling every larger portion “wrong”. People eat differently, and context matters. A snack, a breakfast, a family meal and a shared treat are not the same situation.

The useful question is simpler:

Does the serving information help people understand what they are actually likely to eat?

What we do not share

PerBiscuit does not share names, contact details or individual profiles.

Any industry-facing analysis should use anonymous, aggregated information only. It should also include the basics that make evidence useful:

• Sample size
• Collection period
• Category and product scope
• Calculation method
• Assumptions
• Limitations
• Whether the findings are early signals or stronger evidence

A small number of reports can suggest a useful question. A larger, better-tested dataset can support a stronger finding. PerBiscuit should be clear about the difference.

Questions we can explore together

PerBiscuit can work with manufacturers, retailers, researchers and public health organisations on practical questions such as:

• How do reported portions compare with the serving size shown on the label?
• Do people understand the serving guidance when using the product normally?
• Do pack sizes, resealable formats or multi-serve packs affect portion size?
• Would clearer wording, diagrams or measuring guidance help?
• Are there products where the labelled serving appears technically compliant but practically unrealistic?
• Are there products that deserve credit for realistic portion guidance?
• Could nutrition information be shown more clearly per likely portion, not only per labelled serving?

The strongest projects will focus on specific, answerable questions rather than broad claims about whether a product is “good” or “bad”.

How collaboration could work

1. Agree the question

The partner and PerBiscuit define a useful question before looking at results.

For example:

Do consumers usually pour more than the labelled serving of this cereal?

That is a better question than:

Is this cereal misleading?

Specific questions produce better evidence.

2. Bring the information together

A manufacturer can clarify product details such as pack size, serving guidance, formulation, intended use and labelling constraints.

PerBiscuit can contribute eligible, anonymous, aggregated portion observations.

Those two views can then be compared without pretending that either one tells the whole story on its own.

3. Check the method

Where appropriate, researchers, nutrition specialists or public health organisations can help test the assumptions.

This matters because portion data can be messy. People estimate. Products vary. Households use food in different ways. Good analysis should not hide that. It should explain it.

4. Learn and improve

Findings could inform:

• Clearer serving guidance
• Better pack-front explanations
• Improved portion illustrations
• Product development
• Nutrition communication
• Category-level research
• Consumer education

Sometimes the useful result will be a packaging change. Sometimes it will be a better explanation. Sometimes it will simply be a clearer understanding of what the current evidence can and cannot show.

5. Share what is useful

Where possible, PerBiscuit wants general findings, methods and lessons to be published.

That does not mean exposing private commercial information. It means sharing enough for consumers, researchers and other organisations to learn from the work.

An example project

A cereal manufacturer might share up-to-date information about its pack formats, serving guidance and the reasoning behind its labelled portion.

PerBiscuit could compare that with a sufficiently large set of anonymous portion reports from people using similar products at home.

An independent researcher could then review the method, check the assumptions and help interpret the results.

The project might ask whether consumers understand the serving size, whether the pack format encourages larger portions, or whether a clearer bowl illustration would help.

A useful result could be:

• A better on-pack portion example
• Clearer wording
• A new research question
• A category-level insight
• Recognition that the existing serving guidance is already realistic
• A public explanation of the limits of the available data

This would be a shared learning exercise, not a paid judgement on the company or on individual consumers.

Principles for working together

Privacy

Use anonymous, aggregated data. Do not expose individual reports, names, contact details or profiles.

Clarity

Keep observed data, calculations, assumptions and opinions separate.

Independence

Partners may correct factual errors, explain context and challenge methods. They do not buy a favourable result.

Proportion

Do not make claims that go beyond the size or quality of the evidence.

Practical value

Focus on changes that make serving sizes, portions and nutrition information easier for people to understand.

Shared benefit

Consumers should get clearer information. Companies should get useful feedback. Researchers and public health organisations should get better questions and better data.

Why this matters

Serving sizes are not just numbers on a label. They shape how people compare products, understand calories, read nutrition information and judge what they are eating.

When the labelled serving and the real-world portion are far apart, the label can become less useful, even when it follows the rules.

PerBiscuit wants to help close that gap.

Not by blaming people for eating normally.

Not by attacking companies for every imperfect label.

But by making portion information clearer, more realistic and more useful in everyday life.