Neuromarketing

Sensorimotor, cognitive, and affective reactions are measured: eye movements, electrical brain activity, and physical arousal. These values are recorded while a test subject perceives advertising, packaging, prices, or user interfaces.

The results mostly serve questions regarding advertising effectiveness, packaging, brand perception, price presentation, and user experience. The term stands primarily for commercial application. Academic research at the same intersection, however, is usually called consumer neuroscience.

The methods record physical signals and statistically correlate them with the stimulus shown. A direct look into thoughts or purchase intentions is not possible with this approach, because what is always measured is only a correlate, such as blood flow in the brain or skin conductance.

The term refers to three different things: the measurement practice using devices, academic consumer neuroscience, and a popular label. Under this label, marketing recommendations are also sold in the DACH region without any measurement of their own. The dispute extends into the academic literature. Lee et al. understand neuromarketing as an independent scientific field. Hubert and Kenning, on the other hand, classify it as a business activity and propose the term consumer neuroscience for research.

What is actually measured in the laboratory

There are only a few methods, and each captures a nameable physical or physiological quantity. Functional magnetic resonance imaging (fMRI) measures blood flow in the brain and derives activity patterns in individual brain regions from it. EEG and MEG record electrical or magnetic signals from the cerebral cortex, respectively. Eyetracking records gaze positions and fixation durations. It therefore measures attention and explicitly not brain activity, although many portrayals equate it with the overall field.

In addition, there are biometric methods that do not rely on brain measurements. Electrodermal activity (skin conductance), heart rate, and respiratory rate reflect physiological arousal. Automatic facial expression analysis classifies facial movements into emotional categories. The key principle for interpretation is that arousal is not directional. An increase in skin conductance indicates that a stimulus has been activated, not whether it is pleasant.

None of these methods directly measure a purchasing decision. Anyone evaluating results therefore first checks which metric is in the report, and then which interpretation step derives a statement about brand impact or purchase intent from it. This is where studies diverge.

fMRI or EEG?

The choice between imaging techniques is determined by the trade-off between spatial and temporal resolution. fMRI is the most common imaging method in neuromarketing. It allows localization in the range of a few cubic millimeters, even in deeper structures such as the amygdala and nucleus accumbens. The price for this is a sluggish signal: the images summarize around ten seconds of activity.

EEG and MEG reverse this relationship. They have finer temporal resolution than fMRI. Their spatial resolution has only recently been improved through signal processing and remains inferior to hemodynamic measurements. This leads to a rule of thumb: When determining which brain region is involved, there is no substitute for fMRI, especially for deep structures. When determining where within a specific area the response begins, however, EEG or MEG are better suited due to their fine temporal resolution. Added to this is the cost, as fMRI studies are considered the most expensive method.

What eye tracking answers

Eye tracking is the most well-validated tool for packaging design, shelf placement, and UX issues. It reveals what is seen, in what order, and for how long. For example, it identifies elements on packaging that are overlooked.

The most common misinterpretation is when someone infers a preference or purchase intent from a long fixation. Numerous studies show that long fixation duration and the final choice made diverge. After all, a gaze can also linger because an element is confusing or difficult to decipher. Eyetracking should therefore be used for what it measures, namely visual attention. Whether a variant is also chosen is only revealed by additional behavioral data.

Does this work better than a survey?

The effect of brand knowledge, which systematically escapes surveys, is best documented. In 2004, McClure and colleagues gave 67 volunteers Coca-Cola and Pepsi in *Neuron*, once anonymously and once with brand branding. They combined behavioral taste tests with passive fMRI experiments. In the anonymous condition, a consistent neural response was observed in the ventromedial prefrontal cortex. As soon as the brand was known, brand knowledge for one of the two drinks influenced the stated preferences and the measured brain responses. This Pepsi paradox laid the foundation for the discipline. It shows that brand impact begins at a level that cannot be revealed by information provided in a taste test.

Neuroforecasting is the established approach to forecasting. Knutson and Genevsky define it as the use of neural activity to predict aggregated market decisions. It is therefore about the behavior of many, not the purchase of a single individual. This outlines the scope of the statement. Neurophysiological data can supplement surveys where the latter cannot capture implicit processes, and they can contribute to forecasting at the aggregate level. However, they do not provide individual purchase predictions. Reports that suggest otherwise exceed the proven explanatory power of these methods.

The buy button that doesn't exist

The idea that brain scans could reveal a trigger for purchasing behavior has been associated with the term from the very beginning. There is no scientific evidence for such a „buy button.“ The brain processes behind a purchasing decision are too complex for any clear-cut triggers to have been identified. Even providers themselves acknowledge this and openly admit that they cannot trigger the “buy button” in the customer’s mind.

The manipulation allegation thus collapses due to the limited predictive power. Consumer advocates fear that the procedures are being deliberately misused for influence. So far, however, the measurements do not provide any guidance for control, but only descriptions of reactions. For the selection of providers, this results in a testing criterion: anyone who sells a measurement as access to unconscious purchase triggers is promising more than the data supports.

Added to this is a conceptual ambiguity. In a sense, every marketing is neuromarketing, because campaigns are almost always intended to generate brain activity that leads to the desired behavior. The label therefore does not draw a sharp distinction between measured practice and conventional advertising effect research.

What GDPR and the EU AI Act permit

Two regulations form the framework. Measurement data such as brain activity, skin conductance, and facial images involve biometric and health-related data. Article 9 of the GDPR treats them as a special category of personal data and thus subjects them to stricter requirements. In addition, Article 5 of the EU AI Regulation prohibits manipulative techniques. This does not preclude procedures that measure and document reactions. However, it does apply to systems designed to derive targeted influence from such data.

To be distinguished from this is subliminal advertising, with which neuromarketing is regularly confused. It is legally regulated independently and describes a different set of circumstances: it introduces stimuli below the threshold of perception and does not measure reactions to visible stimuli.

How consent, purpose limitation, anonymization, and retention of measurement data are specifically regulated in the DACH region cannot be reliably deduced from the available sources. [Research gap, human review]

When a neuromarketing study is worthwhile

The application depends on whether the question targets a quantity that one of the methods measures. Three cases speak in favor of this:

• When it comes to attention-related questions regarding packaging, shelf placement, and user interfaces, eye tracking provides direct answers because it is best validated in these areas.
• In cases of implicit brand effects that surveys fail to capture, neurophysiological measurements modeled after the Pepsi Paradox can provide additional insight.
• For forecasts at the aggregate level, neuroforecasting is the methodologically defined approach.

Equally clear are the cases where the effort does not pay off. Those who want to predict the purchase probability of individuals are asking for something that these methods cannot deliver. Those who want to test one design variant against another and can observe actual behavior will get a more definitive answer from an A/B test. And behavioral economic effects like anchoring or framing can be tested in everyday experiments, just as Dan Ariely conducts them without fMRI or EEG.

In terms of cost, the order of the methods serves as a guide: eye-tracking and biometric methods are relatively affordable, while fMRI is considered the most expensive method. There are no reliable price ranges available for studies in the DACH region. [Research gap, human testing]

Neuromarketing, Consumer Neuroscience, or Behavioral Economics?

The dividing line runs between science and contract research. Plassmann, Ramsøy, and Milosavljevic classify consumer neuroscience as academic research at the intersection of neuroscience and consumer psychology. Neuromarketing, on the other hand, refers to the commercial interest in neurophysiological tools for the market research of individual companies. In the Gabler Wirtschaftslexikon, Peter Kenning defines consumer neuroscience as a research approach that systematically integrates methods, findings, and theories of neuroscience into marketing.

The neighboring terms can follow this:

• Neuroeconomics is the umbrella term for neuroscientific decision-making research, including fundamental economic questions, as discussed, for example, by Kenning and Plassmann (2005). Marketing is just one application of this field.
• Neuroforecasting is the narrowly defined subfield that uses neural activity to predict aggregated market decisions, not to forecast individual purchases.
• Marketing and consumer psychology as well as behavioral economics They use surveys, observation, and experiments, but do not involve physiological measurements. Anchoring effects or framing are therefore not neuromarketing in the strict sense.
• Limbic® It is a commercial segmentation model and explicitly not a measurement method. Hans-Georg Häusel developed it in the late 1990s with the Nymphenburg Group. Experts criticize it for reducing the complexity of the brain to three basic systems and a few terms.
• Neuroselling applies the same principle to the sales pitch and the purchasing decision in direct contact.

In practice, this means that a solution that uses a segmentation model doesn't measure anything. This distinction determines whether the end result is actual data or merely an interpretive framework.

Figures that should not be cited

Three statements regarding the term circulate so frequently that they appear to be established without actually being so.

The first is the claim that 95 percent of all purchasing decisions are made unconsciously. Dozens of German-language sources cite this as a fact and pass it on. A verifiable primary source with a study design and sample cannot be reconstructed from this. [Research gap, human review]

The second is the claim that EEG-supported machine learning systems predict purchase intent with an 87.1 percent accuracy. It comes from a market research report without study details, sample size, or test design. Even if such a classification rate was achieved in the laboratory, it cannot be transferred to market forecasts.

The third concerns the market volume. For 2025, Mordor Intelligence cites 1.71 billion US dollars, Research and Markets 3.33 billion, and Market Research Future 27.26 billion. The figures diverge by more than a factor of 10, and none of them disclose their methodology. The only reliable conclusion that can be drawn from this is that dependable market figures for the field are lacking. The widely cited figure of more than 150 commercially active consumer neuroscience companies should be treated with similar caution. It dates back to a review with data from around 2020.

FAQ