Five neuromarketing myths that keep good brands from ever testing.

Hans Berger recorded the first human electroencephalogram in 1924. A century of refinement later, the technology still measures the same basic thing: voltage changes at the scalp, produced by large populations of neurons firing together. It has never, at any stage of that century, produced a device that decodes a specific thought. That gap, between what the instrument has always measured and what people assume it now delivers, explains more resistance to neuromarketing than any real limit in the underlying science does.

We hear the same handful of objections from marketing heads and finance directors weighing a first study, whether the brand sells rum in Kingston or insurance in Port of Spain. Below are the five that come up most, checked one at a time against what the field does rather than against its own reputation.

The myth that it reads minds

This is the myth we hear first, usually within five minutes of a new client call. Someone pictures a cap wired to a screen that prints out a sentence, and assumes we are selling something closer to interrogation than research. We are not, and nobody legitimately operating in this field is either.

What an EEG cap records is voltage at the scalp, produced by large populations of neurons firing in sync. From that signal, trained analysts derive four broad measures: attention, arousal, cognitive load and emotional valence, the last one covering whether a response leans positive or negative and roughly how strongly. What a webcam facial read records is the movement of facial muscles, classified against categories such as joy or disgust. Neither signal carries the content of a thought. Neither can distinguish a want rooted in nostalgia from a want rooted in nothing more than a matching colour scheme. The equipment cannot tell them apart, and nothing sold under the name neuromarketing currently can, regardless of what a vendor's own deck implies. Our buyer's guide to EEG, eye tracking and facial coding covers each signal in more depth, including where it holds up and where it does not.

The number that does the most damage here is the invented one: a report stating "purchase intent: 73 percent" as though a single figure fell out of a brainwave. Raw neural and facial signals do not resolve to a purchase-intent percentage on their own. That kind of number, when it appears, has been built by layering a statistical model on top of the raw signal and calibrating it against past sales data from a specific category. Strip the model away and the raw output is a set of curves: attention rising and falling across a video, and valence tilting positive or negative at specific seconds. That is what the signal supports. A percentage claiming to read intent is doing work the electrodes themselves never did.

The myth that it is only for huge budgets

The image most people carry is the one film and television reach for in any brain-scanning scene: a hospital-grade machine producing a printout that costs six figures. A little of the field's work looks exactly like that. What brands commission in practice is usually smaller and considerably cheaper than the image suggests.

A single-asset pre-test, one ad run through webcam-based eye tracking and facial coding, sits in a completely different price tier from a full multi-signal lab study built around a brand-equity tracker. Pricing the two as though they were one purchase is where the huge-budget assumption comes from. Cost scales with the size of the question being asked: a brand choosing between two cuts of a thirty-second ad before a media buy is asking a small, specific question, and a holding company rebuilding brand architecture across several markets is asking a much larger one. Sample size is the other lever inside a single tier: a directional read from thirty respondents costs less and moves faster than a statistically powered read from a hundred and fifty, and either can be the right call depending on what the decision is worth. Orbital prices each accordingly. A tailored brief lands within 24 hours of a request, and a full study cycle runs three to five weeks depending on scope and sample size, with study designs available from a single-ad pre-test up to full brand-equity tracking.

The practical test for a brand unsure where it falls: bring the smallest, most specific question available (a single ad, or a single pack redesign) and ask for a quote against that question alone. The number that comes back rarely resembles the industry's own reputation.

The myth that it is just a fancier focus group

Wire a respondent up to sensors and ask what they thought of the ad: to a sceptical marketing director, that can look like a focus group with better production values. The equipment does not change what happens if a study still ends with a question and a spoken answer. Self-report stays self-report, no matter how much hardware is in the room with it.

The real distinction is where the value sits. A focus group's output is what a respondent says out loud, filtered through memory and the pull to sound like a coherent person. We wrote at length about how badly that filter distorts the data in why surveys lie. Neuromarketing's output is the response that never reaches language: the half-second attention dip before a respondent consciously registers what they missed, and the flicker of disgust corrected off the face before it becomes a comment. Pupil dilation shows the same thing happening lower down again, a spike in arousal with no vocabulary attached to it at all. None of that gets captured by a facilitator with a clipboard, however skilled the facilitator. It gets captured by instruments built to catch a signal before the speaking mind has a chance to edit it.

Self-report stays self-report, no matter how much hardware is in the room with it.

A well-run study still asks the verbal question, because the answer is a useful cross-check against the measured response. The two frequently disagree, and when they do, the measured response is the one that tends to predict what happens in market.

The myth that it is unethical or invasive

This objection deserves a straight answer, because the underlying worry is reasonable. Reading a person's brain activity or facial response sounds like a bigger intrusion than handing them a clipboard, and for a field with a history of overselling itself, scepticism about intent is fair.

Here is the standard we operate to, and the standard any legitimate practice in this field should be held to. Every participant gives explicit informed consent before a session begins and can withdraw at any point without giving a reason. Data is stripped of identifying detail before analysis and stored securely, in full compliance with the Jamaican Data Protection Act 2020 and the equivalent data protection laws across the other Caribbean markets we work in. No participant's raw facial video or EEG trace leaves the lab attached to their name. What a client receives is aggregated findings: curves and patterns across a sample.

This is the operating baseline we hold ourselves to on every study, matching the standard that governs legitimate biometric and health-adjacent research generally. Getting this wrong costs a brand more than a compliance headache: a study built on carelessly handled participant data becomes the brand's own reputational exposure the moment anyone asks how the numbers were produced. Any neuromarketing vendor should be able to state their version of it before a session gets booked. Full detail on how Orbital handles consent and data sits in our FAQ.

The myth that it is too slow for a real campaign timeline

Campaign timelines are usually the real constraint behind a no, more than budget or scepticism about the science. A media plan is locked and a launch date is fixed; nobody wants to introduce a research process that might blow through either. The assumption is that anything involving a lab and trained analysts must run on a slower clock than a normal marketing sprint.

In practice, a tailored research brief comes back within 24 hours of a request, and a full study cycle, covering recruitment, lab sessions, data collection and analysis, runs three to five weeks depending on scope and sample size, ending in recommendations ranked by expected impact. That timeline compares well against the process it usually gets weighed against: a traditional multi-market survey wave. Fielding a survey across several Caribbean markets, waiting for quota fill, chasing incomplete responses and cleaning the data can itself run months, especially once a client factors in a second wave to confirm a shaky first result. A poorly run survey process is rarely the fast option it gets assumed to be, and it is worth timing both properly before deciding which one threatens the launch date.

The honest caveat: three to five weeks is still not zero weeks, and a study started the week before a launch is a study started too late. The fix is scheduling. Building a pre-test into the plan at the same point a media buy gets locked leaves comfortable room for the result to change the creative before it airs.

Where today's scepticism comes from

None of today's scepticism was invented from nothing, and it is worth saying so rather than treating every objection as a misunderstanding waiting to be corrected. In the early 2000s, as the field was still finding its name, some firms marketed the idea of a literal buy button in the brain: a single region of neural activity that, once located, would predict and even trigger a purchase decision on command. The underlying neuroscience never supported that claim, and the overreach did real damage to the field's credibility, damage that a marketing director weighing a first study today is still, reasonably, carrying some memory of. Two decades of research since have settled on a narrower and better supported claim: attention, arousal, cognitive load and valence can be measured directly, and that measurement predicts behaviour more reliably than asking people to explain themselves. That narrower claim is the one modern neuromarketing makes. Professional bodies and peer-reviewed replication took the place of press-release claims, slowly, over the two decades in between. The scepticism inherited from the buy-button era is a fair inheritance. The methods sold under the name today are a different, more modest thing than what was sold then.

The fastest way to find out which of these five does not apply to a brand is to run one small test against a real decision. Bring Orbital the question at orbitalbrandscience.com, and a tailored brief comes back within a day.

Frequently asked questions

Is neuromarketing regulated or overseen by any ethical standard?

There is no single global regulator built specifically for neuromarketing, and that gap is part of why the ethics question keeps coming up. What governs the field is a layer of existing frameworks: data protection law such as Jamaica's Data Protection Act 2020, and general human-subjects research ethics such as informed consent and the right to withdraw at any time. Industry bodies including the Neuromarketing Science and Business Association also maintain a code of conduct for member firms. A credible provider operates inside all of those layers at once, rather than waiting for a neuromarketing-specific law that does not yet exist.

What can neuromarketing tools actually not tell you?

Neuromarketing tools measure response. They will show that attention dropped at second eleven of an advert, or that a pack redesign produced a stronger positive facial response than the version it replaced. Turning that measurement into a strategic recommendation stays a human, interpretive job the tools do not do on their own. They also do not replace basic stated-fact research: a survey answers questions about brand awareness or self-described usage faster and more cheaply than a biometric study runs. Most of the signals need a controlled setting to stay reliable. EEG in particular loses accuracy outside a lab, which limits how well it travels into a busy retail environment.

How much does a first, small neuromarketing test typically cost to try?

There is no fixed public price list, because cost is scoped to the decision. The honest guidance for a first-time client is to start as small as the question allows. A single-asset pre-test on one piece of creative, run through eye tracking and facial coding rather than a full multi-signal build, is the standard entry point, priced and scheduled independently of any larger tracking programme. Orbital quotes within 24 hours of a brief, so the answer to how much arrives faster than most clients expect.

What happens to the biometric or facial data collected from participants?

Raw facial video and EEG data are stripped of identifying detail before they enter analysis, so an individual participant's data is disconnected from their name and contact details as early in the process as the study design allows. The material is stored securely and used only to build the aggregated findings a client receives: curves and patterns across a sample. Clients do not get raw footage or an individual participant's file. Participants consent to the collection in advance and can withdraw at any point, which removes their data from the study.

Where did the scepticism about neuromarketing originally come from?

The scepticism largely originated in the industry's own early marketing. In the early 2000s, several neuromarketing firms promoted the idea of locating a literal buy button in the brain, a single region of neural activity that would predict or even trigger a purchase decision. That claim outran the neuroscience available at the time and was never borne out by later research, and it did lasting damage to the field's credibility. Present-day scepticism about mind-reading claims or manipulation is, in large part, an inherited reaction to that era rather than a response to what current practice claims or delivers.