Best Neuromarketing Tools and Software
Six kinds of measurement are sold under one label. Each records one signal, measures some things well and others badly, and can be judged with the same few questions.
Answer to the question
What Is the Best Neuromarketing Software?
Neuromarketing software falls into six categories: eye tracking, EEG, facial expression coding, galvanic skin response, implicit association tests and online behavioural testing. Each records one signal. The best choice is the tool whose signal matches your question, whose method is reported in full and whose gaps a second measure covers.
Key takeaways
- Neuromarketing software covers six kinds of measurement, and each records one signal: gaze, scalp electricity, facial movement, skin conductance, reaction time or online behaviour.
- In the 2001 picture-viewing study by Margaret Bradley and colleagues, skin conductance responses varied with arousal and were largest for the most arousing pictures, both unpleasant and pleasant, so the signal marks intensity and not direction.
- The 2019 review by Lisa Feldman Barrett and colleagues found that facial movements communicate useful information but not emotion in a simple one-to-one way, so movement counts are sturdier than emotion labels.
- The 2009 meta-analysis by Anthony Greenwald and colleagues covered 122 reports and found that implicit tests and self-report each predicted outcomes and each added something beyond the other.
- Lee Cronbach and Paul Meehl introduced construct validity in 1955, the question of whether a score measures the quality its label names, which is the first question to put to a vendor.
Neuromarketing software is sold as one product category, yet it covers six kinds of measurement that record different things. Martin Lindstrom's Buyology argues that what people say about why they buy is unreliable and that measurement can get beneath it.
Which six kinds of neuromarketing software are there?
Each category answers that argument in its own way and leaves part of the question open. A buyer who knows which signal a tool records, and what the vendor infers from it, can compare tools without a ranking or a brand list. The table sorts neuromarketing software into six categories by what each records, because the category fixes what a tool can tell you whoever makes it.
| Method | What it records | Measures well | Measures badly | Question it suits |
|---|---|---|---|---|
| Eye tracking | Where the gaze lands, in what order, for how long | Attention to parts of a page, pack, screen or shelf | Liking, memory and intent to buy | Does the key element get seen? |
| EEG | Electrical activity at the scalp, moment by moment | Shifts in response across the length of a video | Locating brain regions; movement and noise disturb it | Where in an ad does the response change? |
| Facial expression coding | Movements of the face, from camera footage | Whether a smile or brow raise occurred, and when | Which emotion a movement means | Do viewers react visibly at this scene? |
| Galvanic skin response | Small changes in how the skin conducts a faint current | Level of arousal | Whether the feeling was pleasant, and its cause | Which moments raise arousal? |
| Implicit association tests | Speed of sorting words and images under paired categories | Strength of a mental link between a brand and a trait | Forecasting one person's choice | What associations does a brand hold in a group? |
| Online behavioural testing | Clicks and choices from randomly assigned versions | What people do on a live page, at scale | Why they did it | Which version of this page works better? |
What can each method show, and what can it not?
Eye tracking in neuromarketing software records where the gaze lands and for how long, and it shows attention in the plainest form of any method here. It suits packaging, shelf layouts, web pages and the order in which the parts of an advertisement are seen. It does not say whether the viewer liked what they saw, remembered it or meant to buy it, so the gaze record needs a second measure beside it. Eye Tracking in Neuromarketing: What Customers Really See covers the method in detail.
EEG records electrical activity at the scalp through a cap of sensors, and it follows a response from moment to moment. That timing suits a video, where a team wants to see which scene shifts the response. The same sensitivity picks up head movement, blinks and muscle tension, so recordings are cleaned before analysis, and the tool says little about which brain region produced a signal. EEG Neuromarketing: Measuring Emotion in Ads gives the method its own article.
Galvanic skin response, also called electrodermal activity, tracks small changes in how well the skin conducts a faint electrical current. In a picture-viewing study by Margaret Bradley and colleagues, participants looked at images with varied emotional and neutral content while several responses were recorded. Skin conductance responses varied with arousal and were largest for the pictures rated most arousing, which included both unpleasant and pleasant content, while facial patterns were specific to the content. A rise in skin conductance marks that something stirred the viewer. It does not mark whether the feeling was welcome.
Facial expression coding in neuromarketing software reads movements of the face from camera footage and counts them, such as a raised brow or a smile. Counting the movement is the part the method does well. Turning the movement into a named emotion is the part that carries risk.
The 2019 review by Lisa Feldman Barrett and colleagues found that people sometimes produce the expected facial configurations, that expressions vary across cultures, situations and individuals, and that the same configuration can carry different meanings. The authors concluded that facial movements communicate useful information but not emotion in a simple one-to-one way. A buyer can use the movement counts as a record of visible reaction and treat any emotion label as a claim to test. What a Camera Can Read on a Face goes further into that research.
Implicit association tests time how quickly a person sorts words and images when two categories share a response key, and the speed pattern is read as the strength of a mental link, for instance between a brand and a trait. A meta-analysis by Anthony Greenwald and colleagues covered 122 research reports and found that such tests predicted behavioural, judgement and physiological outcomes. Self-report measures predicted outcomes too, each kind of measure added something beyond the other, and self-report was weaker on socially sensitive topics.
For brand work the sensible use is a pairing of the two, read as a picture of associations in a group. Does the implicit association test work looks at how steady the scores are for one person.
Online behavioural testing sits at the other end of the scale from the sensor methods in neuromarketing software. It randomly assigns visitors to versions of a page, an advertisement or an email and records what each group does, so the measure is behaviour in the real setting. It answers which version performed better and is silent on the reason, which a sensor signal or a survey answer has to supply separately. What Does an AB Test Actually Prove sets out what such a test can establish.
Judging a tool and its claims before you buy
Lee Cronbach and Paul Meehl introduced the idea of construct validity in 1955, after a committee on psychological tests had found it necessary to distinguish four types of validity, each established by different research. Construct validity is the question of whether a test measures the quality its name claims. Put it to any neuromarketing software that reports a score called engagement, attention, emotion or desire: what evidence shows that the score tracks the quality in its label?
Reporting is the second test. In 2012 a committee of the Society for Psychophysiological Research published recommendations for electrodermal data that cover three recording techniques and the methodological details a report should state. The same habit can be asked of any sensor-based neuromarketing software: which recording method, which settings, which cleaning steps, and how many participants of what kind.
Five questions turn those two tests into a purchase conversation. Which single signal does the tool record, and which claims does the vendor draw from it? What is the score compared against, and where can that comparison be inspected? Who were the participants, and do they resemble your buyers? What does the tool leave out, and which second measure covers the gap? What will a study cost in participants, analysis time and licence terms, written down as a scope before work starts?
The best fit for a given team is the one whose signal matches its question, whose method is reported in full and whose gaps are covered by a second measure. A team that wants to know whether shoppers notice a pack can begin with gaze. Then it can add a memory question and a live test.
Common questions
What is the best neuromarketing software?
No single product is best for every question. The best fit records the signal your question needs, reports its method in full and is paired with a second measure that covers what it cannot see.
Does neuromarketing software read minds?
No. Each tool records a physical or behavioural signal, such as gaze, skin response, facial movement or clicks, and a person or a model interprets it. The interpretation is the part to test.
Can one tool cover every method?
A single platform may bundle several signal feeds, but each feed keeps its own strengths and limits. Check each one against the question you are asking.
What drives the cost of neuromarketing software?
Hardware, participant numbers, analysis time and licence terms drive it, and quotes differ by supplier, so no figure is given here. Ask for a written scope.
Sources
Lindstrom, M. Buyology. Seed book; vault hub Books/Buyology.
Bradley, M. M., Codispoti, M., Cuthbert, B. N., & Lang, P. J. (2001). “Emotion and motivation I: Defensive and appetitive reactions in picture processing.” Emotion, 1(3), 276-298. doi:10.1037/1528-3542.1.3.276 https://doi.org/10.1037/1528-3542.1.3.276
Barrett, L. F., Adolphs, R., Marsella, S., Martinez, A. M., & Pollak, S. D. (2019). “Emotional Expressions Reconsidered: Challenges to Inferring Emotion From Human Facial Movements.” Psychological Science in the Public Interest, 20(1), 1-68. doi:10.1177/1529100619832930 https://doi.org/10.1177/1529100619832930
Greenwald, A. G., Poehlman, T. A., Uhlmann, E. L., & Banaji, M. R. (2009). “Understanding and using the Implicit Association Test: III. Meta-analysis of predictive validity.” Journal of Personality and Social Psychology, 97(1), 17-41. doi:10.1037/a0015575 https://doi.org/10.1037/a0015575
Cronbach, L. J., & Meehl, P. E. (1955). “Construct validity in psychological tests.” Psychological Bulletin, 52(4), 281-302. doi:10.1037/h0040957 https://doi.org/10.1037/h0040957
Society for Psychophysiological Research Ad Hoc Committee on Electrodermal Measures (2012). “Publication recommendations for electrodermal measurements.” Psychophysiology, 49(8), 1017-1034. doi:10.1111/j.1469-8986.2012.01384.x https://doi.org/10.1111/j.1469-8986.2012.01384.x