Are the marks even noticed?

Recognition is not attention

Before a label can inform or instruct, it has to be seen. Two measurements are often mixed together. Recognition is whether a consumer, shown the mark, says they have seen it before. Attention is whether their gaze lands on it at the shelf or the bin. Recognition figures are high and rising. They are the figures schemes publish. Attention figures are substantially lower. They come from eye-tracking work that schemes cite less often.

A mark can score high prompted recognition and still receive little or no fixation at the moment of decision. Prompted-recognition surveys cannot detect that outcome.

Eye-tracking

Ruppenthal (2023) reviewed 38 eye-tracking studies on sustainable food consumption. In the review’s words: “sustainable labels are hardly noticed by consumers in competition with other product information.” “Food product appearance, price, and nutritional information are often fixated first and longest. Eco-labels are neither the first nor the longest viewed AOIs [areas of interest].” Brand logos and images are “more concise, larger, and centrally positioned than sustainability-related elements” (Ruppenthal, 2023). A majority of participants (54 per cent) “do not evaluate the product information of the foods they purchase, suggesting habitual shopping.” Strength grade: moderate-to-strong as a synthesis. Most underlying studies are laboratory eye-tracking, with the ecological-validity limits that implies, but the direction is consistent across 38 studies and matches pack geometry: sustainability marks are small, peripheral and numerous. Fixation follows size, centrality and relevance.

Where gaze does land on an eco-label, it predicts choice: “the longer consumers gaze on foods with organic labels… the more likely they are to choose that product” (Ruppenthal, 2023). Attention is scarce, not useless. The constraint is capture, not processing. That distinction bears on design.

WRAP found most consumers spending no more than ten seconds examining on-pack labelling (WRAP, 2020). The figure comes from a subsample of 30, so the grade is weak, but the direction matches the eye-tracking record. Ten seconds is the interval in which every disposal instruction and certification seal has to compete with brand, claims and mandatory information.

Reported use

Between attention and recognition sits reported use: whether consumers say they consult sustainability labels. Grunert, Hieke and Wills (2014) surveyed 4,408 respondents in six European countries on four marks — Fair Trade, Rainforest Alliance, Carbon Footprint and Animal Welfare. “Sustainability labels currently do not play a major role in consumers’ food choices.” Measured use was low however it was asked (Grunert et al., 2014). The instrument recorded self-reported consultation, not observed behaviour at the shelf. Strength grade: moderate — large multi-country sample, but self-report, food only, and 2013 fieldwork. No later multi-country survey of label use at comparable size was found.

Prompted recognition

Recognition sits at a different level, from different sources, with a different method. The figures below are prompted recognition unless stated otherwise.

EU Ecolabel: “38% of EU citizens recognise the EU Ecolabel logo, an increase of 11 percentage points since 2017,” from Flash Eurobarometer 535, more than 25,000 respondents across the EU-27 (European Commission, 2023). Strength grade: strong for prompted recognition from a properly sampled official survey.

Fairtrade: “Three out of four shoppers recognise the Fairtrade label,” and 83 per cent of those who recognise it trust it, from GlobeScan’s 2025 survey of more than 12,900 consumers in 13 countries (Fairtrade International, 2025). Strength grade: moderate — professionally fielded, commissioned by the label owner.

OPRL (UK): Spring 2025 survey of over 5,000 consumers reports 91 per cent recognising OPRL labels, 81 per cent using recycling labels to decide disposal, and 96 per cent knowing what the labels mean (OPRL, 2025). Strength grade: weak-to-moderate. The sample is large, but the survey is self-commissioned. The 96 per cent comprehension figure sits far above every independently measured result in Do consumers understand what the labels mean?

ARL (Australia and New Zealand): awareness of 75 per cent in Australia and 68 per cent in New Zealand in 2021, up from 47 and 44 per cent in 2018 (APCO, 2022). Strength grade: moderate — scheme-published, documented method, repeated waves, so the trend can be read independently of the level.

Major marks achieve substantial prompted recognition, and recognition grows with tenure and promotion. The ARL’s 28-point rise in three years is the largest movement on a repeated instrument in this set. Recognition does not imply noticing at the shelf, use, or comprehension. Prompted recognition is the least demanding measure: the respondent is shown the mark, away from the shelf, and asked only for familiarity. Eye-tracking asks the demanding version of the same question. 91 per cent recognition against labels “hardly noticed… in competition with other product information” is a difference of measures, not a contradiction.

Why the figures diverge

The measures sit at different stages. Prompted recognition measures memory. Attention measures capture at the decision. Use measures self-attributed influence. Each stage loses most of the previous stage’s population. Most shoppers do not inspect product information at all (the 54 per cent habitual-shopping finding). Those who do look at appearance and price first. The sustainability mark receives what remains.

Recognition is also built off-pack, through campaigns, media and ubiquity, not through brief shelf encounters. A mark can accumulate recognition without improving capture at the shelf. The OPRL and Fairtrade figures, set beside Ruppenthal, suggest that has already happened.

The bin is not the shelf. At purchase, dozens of packs compete for seconds. At disposal, one pack is in hand and the question is where it goes. Sorting effects an order of magnitude larger than purchasing effects, in the behaviour article, are consistent with that. Disposal marks are structurally better placed to be noticed than purchase marks, whatever their recognition scores.

Crowded packs

Appearance, price and nutrition take first and longest fixation. Brand elements are larger and more central. Sustainability marks get residual area because, for most purchases, they are residual considerations. A typical multi-market pack then divides that residual among material codes, seals, disposal instructions and system marks. Eye-tracking has mostly measured a single eco-label on an otherwise uncluttered stimulus. Fixation on a real multi-mark retail pack would be lower.

Mandatory marking — harmonised pictograms, compostability statements, deposit marks — adds more claimants. The attention findings predict that mandated marks will capture at the expense of voluntary ones. Recognition surveys will not detect that, because recognition is built off-pack. Moving system-facing information behind a code can free the printed surface for what can be used in a ten-second glance — provided, as the scanning article discusses, nothing essential moves with it.

Recognition advertising raises prompted recognition without altering capture. Size, contrast, position and consistency act on capture. Those factors are constrained by artwork negotiation unless the mark’s position is mandated. On the located evidence, the pack surface allocates a limited number of seconds, most sustainability marks receive few of them, and the measured determinants of capture are geometric rather than promotional.

What this record cannot yet say

Almost all attention evidence concerns food purchasing in laboratories. No eye-tracking study at the bin was found. Recognition surveys are prompted almost without exception. Unprompted recall is scarce. No study linking measured on-pack attention to later disposal by the same consumers was found. The located studies are almost entirely European and Anglophone. Resin codes, compostability seals and deposit marks are essentially unstudied for attention as distinct from comprehension.

Any study that recruits attentive participants — every online experiment, every survey — measures the label after the rarest event in the funnel has already occurred. Field effects will run below experimental effects by roughly the noticing rate. That is why simulated-task results are treated in this section as upper bounds.

The marks receive measurable but limited fixation at purchase. Prompted recognition measures familiarity, not at-shelf capture. The moment that matters most for disposal — pack in hand, bin in front — has not been instrumented. Portable eye-tracking would make such a study feasible. No commissioned example was found.

References

APCO (Australian Packaging Covenant Organisation) (2022) ARL Impact Framework Report, July. Available at: Open source (Accessed: 18 August 2026).

European Commission (2023) EU Ecolabel use and recognition rise, new Commission data shows (Flash Eurobarometer 535), 12 October. Available at: Open source (Accessed: 18 August 2026).

Fairtrade International (2025) Fairtrade remains the go-to ethical label for shoppers (GlobeScan consumer research). Available at: Open source (Accessed: 18 August 2026).

Grunert, K.G., Hieke, S. and Wills, J. (2014) 'Sustainability labels on food products: consumer motivation, understanding and use', Food Policy, 44, pp. 177–189. Available at: Open source (Accessed: 18 August 2026).

OPRL (2025) Consumer behaviour and attitudes towards recycling (Spring 2025 survey). Available at: Open source (Accessed: 18 August 2026).

Ruppenthal, T. (2023) 'Eye-tracking studies on sustainable food consumption: a systematic literature review', Sustainability, 15(23), 16434. Available at: Open source (Accessed: 18 August 2026).

WRAP (2020) On-pack labelling and citizen recycling behaviour. Report mirror available at: Open source (Accessed: 18 August 2026).

Note on sources and verification

Ruppenthal's quotations are verbatim from the open-access review as retrieved. Grunert, Hieke and Wills' sample and conclusions are from the paper's public abstract and preprint page; per-logo recognition percentages sit behind the paywall and are deliberately not quoted. Recognition benchmarks are quoted from the sources cited, with commissioning relationships flagged in the strength grades: the Eurobarometer figure is official survey data; the Fairtrade figure is owner-commissioned GlobeScan research; the OPRL figures are self-commissioned by the scheme, and the 96 per cent comprehension claim is flagged as inconsistent with independent comprehension research; the ARL figures are scheme-published with documented survey waves. The WRAP ten-second figure derives from a subsample of thirty respondents and is graded accordingly. The absence of eye-tracking research at the point of disposal, of unprompted recall data, and of attention-to-disposal linkage studies are absence claims based on searches conducted for this article. The funnel reconciliation is a synthesis of the cited studies, not a finding of any single one of them.

Last verified: 18 August 2026.