Daily intake percentage labels discredited

A new report finds the front-of-pack labelling system is too confusing for consumers to understand.
Learn more
  • Updated:28 Nov 2011

01 .Introduction


A new report shows the Percentage Daily Intake (%DI) front-of-pack labelling system promoted by some food manufacturers fails to provide consumers with the necessary information to easily compare the nutritional content of similar products. 

The report, commissioned by CHOICE and conducted by The George Institute for Global Health (TGI), has found serving sizes vary significantly between similar products in common food categories.

The investigation used TGI's food labels database to examine products across six categories - snack foods, breakfast cereals, cereal and nut bars, ready meals, soups and yoghurts.

Video: CHOICE Better Food Labelling campaign

Front-of-pack labelling explained.

CHOICE has long believed that %DI labels are confusing and require consumers to undertake complex calculations to use the information effectively, and this was confirmed by the independent panel that reported to Australian governments on food labelling in January 2011.

However, this research shows that even if consumers could understand the information on %DI labels, the inconsistent serving sizes on which they are based means they fail to help shoppers compare products.

All products analysed in the survey were available on supermarket shelves in September and October this year.

Serving size variation

In most categories examined there was a wide range of serving sizes for similar products.  All food categories (with the exception of the chilled soup product type) had inconsistent serving sizes and in the most extreme case, the maximum serving size of flavoured and fruit based yoghurts was ten times greater than the minimum.

Products with %DI didn't show any greater consistency in serving size than those without. In some product categories, the research found that the average serving size was lower for products with %DI than products without. 

Corn chips, most breakfast cereals, chilled ready meals and canned soup all recorded significantly lower serving sizes for products with %DI labels. However, the opposite finding resulted from analysis of extruded snacks, frozen ready meals and yoghurt.

What is clear from this research is that vast differences in serving sizes occur across the board and the voluntary adoption of the %DI system by the food industry has not provided consumers with the consistency they need in order to make healthy decisions. 

Within the snack food category, data for 337 products was analysed across eight product types, including corn chips, popcorn and wholegrain chips.

  • Of 40 corn chip products, serving sizes ranged significantly, with a minimum serving size of 25g and a maximum of 100g.  
  • Similarly, 32 popcorn products produced an 87g difference between the minimum (13g) and the maximum (100g).  
  • Potato crisps - the most popular snack food product on record, with 101 products – also produced a significant range between the minimum 19g and maximum 50g serving size. 
  • Within the snack food category, it was interesting to note that similar products from the same brand varied their serving sizes.  For example, Coles Organic Sweet and Salty Popcorn recommended a serving size of just 20g, while Coles Butter Microwave Popcorn had a serving size of 100g. 

In the ready meal category, frozen ready meals, also a popular product with 172 labels, proved to be the most difficult to compare, with the maximum serving size (450g) almost four times that of the minimum (115g). Serving sizes also varied greatly within the well-known brands.  For example Lean Cuisine Classic Beef Stroganoff with Pasta and McCain Healthy Choice Chinese Chicken and Cashew had a 280g serving size, while Lean Cuisine Rich Beef Lasagne was 400g/serve and McCain Healthy Choice Honey Stir-fry Chicken was 420g/serve. 

The variation continued in the ready- to-eat breakfast cereal category.  The 33 plain adult cereal products surveyed produced a five-fold range between the minimum 10g and maximum 55g serving sizes.  Of 110 muesli products, Whisk & Pin Summer Muesli set their serving size at 80g, while the minimum in this category was just 25g.

The only exception to the trend of wide ranging serving sizes was in the chilled soups product category, where 22 products all set the serving size at 300g.

Results confirm CHOICE research

The findings in the George Institute’s new report are consistent with the results of research conducted in 2009 by CHOICE in collaboration with public health groups, including the Cancer Council, which found that %DI labels were not as easy to interpret as traffic light labelling.

While the 790 consumers surveyed initially thought that %DI labelling was easy to use, only 64% of consumers chose the healthier option when using the monochrome %DI table currently used on packaging.  

In contrast, 84% could correctly identify the healthier option when using the front-of-pack traffic light system.


Sign up to our free

Receive FREE email updates of our latest tests, consumer news and CHOICE marketing promotions.


The results, shown in the graphs below, show serving sizes vary significantly between similar products in six common food categories. All products analysed in the survey were available on supermarket shelves in September and October this year.

    For more information, take a look at The George Institute's report.

    Snack foods

    Snack foods

    Breakfast cereals

    Breakfast cerals

    Cereal and nut bars


    Ready meals







    The expert panel that undertook the Review of Food Labelling Law and Policy, commissioned by the Australia and New Zealand Food Regulation Ministerial Council, presented its final report, Labelling Logic, in January 2011.

    The panel studied front-of-pack labelling systems and recognised numerous disadvantages with the %DI system in its final report. The panel found:

    • The system’s reliance on literacy and numeracy skills
    • The significant cognitive processing required to interpret labels may be confusing for some people
    • The panel also recognised inconsistent serving sizes as a major issue with %DI labels.

    On the other hand, the panel noted that there are many advantages of traffic light labelling, including that evidence shows it is the most effective front-of-pack labelling system in terms of facilitating consumer understanding of the nutritional content of food products. 

    The panel also recognised that traffic lights facilitate comparisons between products. Ultimately, the panel recommended: 

    • The development of an interpretive front-of-pack labelling system that is reflective of a comprehensive nutrition policy and agreed public health priorities.
    • The introduction of a multiple traffic lights front-of-pack labelling system. Such a system should be voluntary in the first instance, except where general or high level health claims are made on the label, in which case it should be mandatory.

    Governments around Australia will have a chance to implement these recommendations when health and agriculture ministers from the states and territories, along with the federal and New Zealand governments, meet on December 9 to respond to Labelling Logic.

    What CHOICE wants

    The latest research from the George Institute conclusively confirms what CHOICE has long suspected: that %DI simply does not provide consumers with the information they need to compare products easily and make healthy decisions about what to eat.

    CHOICE supports interpretive front-of-pack labelling, like traffic lights, which give consumers at-a-glance nutritional information and helps inform healthy decisions.

    CHOICE calls on governments to reject the confusing %DI system promoted by industry. Instead, governments need to implement the Labelling Logic recommendations and commit to the introduction of a traffic light-style interpretive front of pack labelling.

    04.Summary: %DI guides vs traffic light labelling


    The current %DI labelling system:

    • Is not standardised, which means consumers who attempt to compare products using the %DI system may be comparing products with serving sizes ranging between 50g and 500g.
    • Is voluntary and therefore not visible on similar products in the same category. 
    • Allows one product to recommend a serving size substantially different to that of a similar product.  Of 28 product types, only one had zero variance. 

    A traffic light system would:

    • Be based on a 100g or 100ml of products, which means consumers could compare products with ease.
    • Interpret the nutritional information and translate it into traffic light representations which can be easily understood at a glance.
    • Be consistent with the recommendations of the expert panel that reviewed food labelling.

    Your say - Choice voice

    Make a Comment

    Members – Sign in on the top right to contribute to comments