What Biometrics Can and Cannot Reveal About Consumers

Researchers monitor a participant connected to sensors and displaying data

Biometric research holds obvious appeal for brand leaders. If a brand lives partly in memory, expectation, recognition, trust, and felt experience, then measures that seem to capture reactions directly from the body can look like a shortcut past self-report. Heart rate, skin conductance, facial coding, and electroencephalography, or EEG, are often presented as ways to reveal what consumers “really” feel when they encounter an ad, package, retail environment, sponsorship, interface, or branded experience.

That promise is worth examining carefully. Biometrics can contribute valuable evidence about attention, arousal, and certain aspects of processing during exposure to brand-related stimuli. They can help researchers understand when something is registering, when intensity changes, or when a sequence produces strain, novelty, or engagement. What they cannot do, at least not reliably or in isolation, is read brand meaning directly from the body or settle complex questions about positioning, trust, perceived quality, authenticity, or long-term brand equity.

For branding professionals, that distinction matters. Brands are not only noticed. They are interpreted. They acquire meaning through repeated exposure, category context, social cues, prior experience, reputation, memory structures, and organizational behavior over time. Biometric data can illuminate parts of that process, but it cannot substitute for strategic interpretation.

What biometric measures actually capture

The first discipline in using biometrics well is to avoid treating all physiological signals as if they measure emotion in the same way.

Heart rate is often used to study changes in attention, effort, orienting, and arousal. Depending on the context and method, increases or decreases may correspond to different kinds of cognitive and emotional processing. A momentary deceleration can be associated with orienting to a stimulus, while other patterns may reflect effort or stress. On its own, heart rate does not tell a researcher whether a consumer likes a brand, distrusts it, feels surprised by it, or is merely concentrating.

Skin conductance, also called electrodermal activity or galvanic skin response, measures changes in sweat gland activity linked to sympathetic nervous system arousal. It is useful for identifying intensity. A spike may indicate excitement, anxiety, surprise, confusion, or tension. It does not identify the valence of that state. A consumer watching a high-stakes brand film, seeing a luxury price point, or confronting a difficult onboarding flow may all show elevated arousal for very different reasons.

Facial coding typically uses computer vision to estimate visible facial movements that may correlate with expressions such as smiling, brow furrowing, eye widening, or disgust-related movements. These systems are often built on variants of the Facial Action Coding System developed by Paul Ekman and Wallace Friesen. Yet visible facial movement is not the same thing as a stable internal emotional state, and the scientific debate around automated emotion recognition is substantial. Context, culture, display rules, individual variation, and the simple fact that people do not always show emotion clearly on their faces all limit what can be concluded. The National Institute of Standards and Technology has also evaluated demographic effects in facial analysis technologies, underscoring that performance and fairness issues remain material research concerns. See nist.gov/programs-projects/face-analysis-technology-evaluation-fate.

EEG records electrical activity from the scalp and is used in consumer research to infer patterns related to attention, cognitive effort, approach-withdrawal tendencies, or memory encoding proxies, depending on the model and experimental design. EEG has excellent temporal resolution, which makes it useful for identifying when processing changes during an ad, experience, or interface journey. But EEG signals are noisy, indirect, and interpretation-heavy. They do not provide a straightforward readout of brand preference or emotional truth.

In short, these tools measure bodily and neural correlates of processing. They do not measure brand equity itself.

Why emotion inference remains difficult

The central problem is not that biometric measures are useless. It is that consumer emotion is multidimensional, context dependent, and entangled with cognition, memory, and meaning.

Arousal is not preference. Attention is not persuasion. Recognition is not trust. Cognitive effort is not necessarily friction, and friction is not always bad branding. A premium spirits label, a luxury retail interior, or a public-interest campaign may intentionally create pause, tension, or reflective effort. If biometric data shows a stronger physiological response in those moments, that may indicate successful strategic differentiation rather than a problem to be optimized away.

This is one reason many broad claims about “real emotion” should be treated cautiously. The Association for Psychological Science and other research communities have repeatedly highlighted the limits of inferring discrete emotions from physiological patterns alone. Similar bodily responses can map to multiple subjective states, and the same stated emotion can appear with different physiological signatures across people and situations.

Facial coding raises additional complications. A smile during exposure to a brand may indicate pleasure, politeness, irony, uncertainty, or social masking. Minimal facial activity may reflect concentration rather than indifference. Consumers often process brands in ways that are quiet, culturally moderated, or cognitively layered. The body does not deliver a clean caption.

For brand strategy, this matters because brands are not judged in laboratories the way they are encountered in life. A heritage brand may evoke reassurance in one audience and staleness in another. A challenger brand’s provocative tone may create productive tension among prospective adopters and rejection among incumbent loyalists. No biometric dataset can interpret those reactions correctly without a broader theory of the audience, category, and brand meaning.

What biometrics can contribute to branding decisions

Used carefully, biometric methods can still be highly relevant to branding. Their value is strongest when the question is specific, bounded, and tied to a real decision.

For example, biometrics can help identify whether distinctive brand assets are actually being noticed within cluttered environments. If a sonic mnemonic, package shape, retail cue, or motion signature is intended to trigger recognition quickly, researchers may use eye tracking, skin conductance, or EEG timing analysis to study whether it interrupts scanning patterns or produces a measurable orienting response. That does not prove the asset strengthens equity, but it can show whether it has a chance to enter memory under realistic conditions.

Similarly, when a brand is refining an experience, biometrics can identify moments of overload or uncertainty that may not surface in recall-based surveys. A complex app onboarding flow, a confusing shelf navigation system, or a loyalty-program sign-up sequence may create elevated cognitive effort or arousal. When paired with observation and follow-up interviews, those measures can help diagnose where the experience conflicts with the intended brand promise.

Biometrics can also be useful in evaluating temporal sequences. Branding often unfolds over time rather than in a single static encounter. A film, event activation, product demo, store journey, audio identity, or unboxing sequence may work because of pacing, escalation, release, and memory peaks. Physiological measures can help show where intensity builds or drops, where attention drifts, and where transitions interrupt processing.

This is most valuable when the brand question is operationally clear. Are consumers registering the parent brand in a portfolio message? Is a sub-brand cue overshadowing the masterbrand? Does a new sonic asset draw attention but misalign with the brand’s intended tone? Does an emotionally charged message create activation without improving attribution? Biometrics can support these inquiries, but only as one form of evidence.

Where brand teams often overreach

The most common error is moving too quickly from physiological response to strategic conclusion.

A high-arousal moment in an ad is not proof of stronger brand positioning. It may reflect category drama, celebrity presence, soundtrack intensity, confusion, or mere novelty. Likewise, increased attention to a package does not necessarily mean improved shelf conversion or stronger distinctive asset recall. It may simply mean the design is unfamiliar or hard to process.

Another overreach is treating biometrics as a replacement for asking people what they think. This is especially problematic in branding because many brand effects are interpretive and longitudinal. Brand trust, perceived quality, purpose credibility, and reputation involve beliefs accumulated across time. They are shaped by product performance, earned media, customer service, pricing decisions, social discourse, and consistency between what a company says and what it does. Those are not reducible to bodily signals during a single exposure.

The same caution applies to rebranding. If an organization updates its identity, architecture, naming system, or messaging, biometric testing may reveal that people respond with surprise, strain, or elevated attention. But those early reactions do not determine whether the rebrand clarifies the portfolio, improves recognition, transfers equity effectively, or better expresses the brand’s strategic position over time. Immediate physiological disruption may be the normal effect of encountering the unfamiliar, not evidence of failure.

This is particularly important because social and symbolic meaning often lags exposure. Consumers may need repeated encounters to interpret what a new identity or naming structure represents. The brand team’s task is not to eliminate every moment of surprise. It is to understand whether the new system ultimately supports clearer recognition, stronger associations, and better strategic coherence.

Biometrics and the difference between advertising response and brand effect

Branding and advertising overlap, but they are not interchangeable. Much biometric research is conducted on advertisements, video content, digital interfaces, or shopper stimuli because those are easier to test experimentally. Yet a strong response to branded communication is not the same thing as a durable change in brand meaning.

An advertisement may drive high attention and emotional intensity while doing little to strengthen memory structures tied to the brand. Conversely, a less dramatic execution may reinforce distinctive assets, category cues, and positioning associations more effectively. This distinction parallels the difference between momentary performance and long-term brand building.

Researchers at the Ehrenberg-Bass Institute, among others, have emphasized the importance of mental availability and distinctive brand assets in making brands easier to notice and buy. See marketingscience.info/assets/documents/Ehrenberg-Bass_Institute_-_How_Brands_Grow_Part_2.pdf. Biometric signals may help analyze whether such assets interrupt clutter or hold attention, but the larger branding question concerns whether those cues become reliably linked in memory to the brand over repeated exposures and buying situations.

That requires measures beyond physiology. Brand attribution, aided and unaided awareness, asset recognition, association mapping, consideration, perceived fit, and longitudinal tracking remain essential. If a biometric study suggests an ad sequence is highly engaging but follow-up testing shows weak brand linkage, brand managers should resist celebrating the wrong outcome.

The interpretive challenge in facial coding and emotion AI

Facial coding deserves special caution because it is often marketed in especially expansive terms. The idea that software can determine what consumers “really feel” by reading expressions is appealing, but many researchers have challenged strong claims about universal and context-free emotion recognition.

A 2019 review in Psychological Science in the Public Interest by Lisa Feldman Barrett and colleagues argued that people do not reliably scowl only when angry, smile only when happy, or show discrete emotional states in uniform ways across contexts. See journals.sagepub.com/doi/10.1177/1529100619832930. That does not mean facial movement is irrelevant. It means inference must be narrower and more careful than many commercial claims suggest.

For branding, this is a practical issue. If a facial coding platform classifies a response to a price reveal as “anger,” is the consumer actually offended, momentarily concentrating, socially signaling, or reacting to something else in the test environment? If a humorous campaign produces smiles, are viewers appreciating the joke, recognizing the celebrity, or enjoying the situation while missing the brand entirely? Without triangulation, the interpretation can become little more than software-assisted projection.

What good practice looks like

Biometrics are most useful when treated as one layer within a mixed-method research design. For brand-related decisions, that usually means combining physiological data with at least some of the following:

  • Behavioral observation, including navigation, gaze patterns, click paths, or shopper movement.
  • Self-report, ideally designed to capture interpretation, not just liking.
  • Recognition and attribution measures, especially when testing distinctive assets or communication linkage.
  • Implicit or associative methods when relevant, though these also require careful interpretation.
  • Brand tracking and longitudinal data to assess whether observed responses relate to durable changes in awareness, consideration, trust, or equity.
  • Qualitative interviews or ethnography to uncover what consumers believed was happening and why.

The key is not methodological excess. It is methodological fit. A brand team testing a sonic identity may need to know whether listeners notice it, connect it to the right brand, and describe it in ways that align with the intended positioning. A biometric signal may help identify salience or intensity at the moment of exposure, but only attribution and interpretation can show whether the asset supports the broader identity system.

The same principle applies to brand architecture. Suppose a company introduces a more unified endorsed system across previously stand-alone products. Biometrics may reveal where consumers hesitate when parsing the new hierarchy, but surveys and interviews are needed to determine whether the architecture improves understanding, equity transfer, and confidence across the portfolio. A pause is only a pause until it is interpreted.

Brand questions biometrics can answer better than most other tools

The strongest use cases tend to involve timing, nonconscious orienting, and moments consumers may not remember accurately afterward.

These include questions such as:

  • When during a branded sequence does attention increase or collapse?
  • Which sensory cues are associated with immediate orienting or arousal?
  • Where in a retail or digital journey do consumers appear overloaded or uncertain?
  • Does a change in packaging, motion, sound, or interface alter the ease with which people register the brand?
  • Are key assets being noticed soon enough to support recognition before attention moves on?

Even here, “noticed” should not be confused with “understood.” A cue can interrupt without communicating the right meaning. A package can stand out while damaging perceived fit with the category. A sonic device can be memorable while feeling inconsistent with the brand’s personality. Branding decisions always require moving from signal detection to meaning diagnosis.

Ethics, privacy, and trust are branding issues too

The use of biometrics also has implications beyond measurement accuracy. It can affect brand trust and reputation.

Collecting physiological or facial data raises legitimate concerns about consent, storage, transparency, secondary use, bias, and consumer comfort. Regulatory scrutiny of biometric information has grown in the United States and internationally. Illinois’ Biometric Information Privacy Act is one prominent example, and other jurisdictions have enacted or proposed related protections. See ilga.gov/legislation/ilcs/ilcs3.asp?ActID=3004.

For brands, the issue is not only legal compliance. It is whether research methods align with the organization’s broader values and promises. A brand that positions itself around trust, care, or human understanding can undermine those claims if consumers or participants feel surveilled or misled. Internal governance around biometric research therefore belongs within brand stewardship as well as insight operations.

That includes being precise in how findings are described internally. Overclaiming what a biometric study “proves” can distort decision-making, but it can also encourage a culture in which brand meaning is treated as something extractable from bodies rather than co-created through relationship and experience. That is bad research and weak brand thinking.

What branding professionals should take from biometric evidence

The most productive stance is neither enthusiasm nor dismissal. Biometrics can sharpen the analysis of branded stimuli, especially in high-speed or high-clutter environments where self-report is incomplete. They can help explain when something happened in the consumer experience, and sometimes how intensely. They are far less reliable at explaining what that moment meant to the person experiencing it or what it will do to the brand over time.

Branding depends on interpretation across repeated encounters. Positioning works when people come to understand a brand relative to alternatives. Identity works when signals become recognizable and meaningful. Reputation develops through experience and organizational conduct. Equity grows when memory, expectation, and perceived value reinforce one another in market behavior. None of those processes can be observed directly through a wrist sensor, camera, or electrode cap.

Biometric research is therefore most valuable when it is used with strategic humility. It can show where to look more closely, where an experience intensifies, where recognition may falter, or where a branded sequence loses people. But the task of branding remains larger: to understand how those moments connect to perception, association, trust, and choice. Physiological data can enrich that understanding. It cannot replace it.

Leave a Reply

Discover more from American Advertising and Marketing Association | AAMA

Subscribe now to keep reading and get access to the full archive.

Continue reading