The Most Dangerous Phrase in Marketing
“We know our customer.”
Four words that have decided more marketing strategies than any framework, focus group, or dashboard. They have also decided on more failed launches than most marketers will ever admit.
The phrase sounds reasonable. The people saying it are paid to understand consumers. They have access to data, judgment, and taste. When they say “we know our customer,” they mean it as a baseline: a starting assumption on which the brief, the positioning, and the creative are built. Over a decade of academic research suggests that the baseline is broken, and the harder marketers try to fix it, the worse it gets.
Why we believe we know our customer
Every part of the marketing process trains us to think we already know what the customer wants. The brief starts with the product. The value proposition is built from the product. The competitive advantage is mapped from the value proposition. The positioning is defined from the competitive read. Each step is internally consistent and inward-facing.
The customer is the last thing the process touches, and by the time we get there, the strategy is already decided. The brain that will actually decode the message is the one place the process never visits.
The result: a deep and well-earned conviction that we know what the customer thinks, when what we actually know is what the company thinks the customer should think. The two are not the same. The research now has names for both.
The bias has a name
The phenomenon is called the false consensus effect. It was first documented in social psychology by Ross, Greene, and House in 1977. People believe others share their preferences more than they actually do. Marketers, despite their training, are not immune.
A research team led by Walter Herzog (WHU, Germany), Johannes Hattula (Imperial College London), and Darren Dahl (UBC Sauder School of Business) has spent over a decade examining this bias specifically within the marketing profession. Their work has been published in the Journal of Marketing Research.
In their 2021 paper, they tested 714 marketing executives across multiple experiments. Even senior, experienced professionals, the ones whose job it is to be calibrated to the audience, confused their personal preferences with their predicted customer preferences. One marketer interviewed for the research, working at an IT firm, summed up the pattern bluntly:
The false consensus effect has a huge impact on many marketing managers in my firm.
The conviction that the room already understands the customer is the operational bias. The more confident the room, the more its decisions reflect the room’s preferences in a different shirt.
Empathy makes it worse, not better
The same research team, this time with Sven Reinecke (University of St. Gallen), had earlier published a more counterintuitive finding. The 2015 paper, “Managerial Empathy Facilitates Egocentric Predictions of Consumer Preferences,” tested the standard advice that every marketer hears: when you want to predict what your customer thinks, take their perspective. Put yourself in their shoes.
The result of four empirical studies with 480 experienced marketing managers, stated in the first sentence of the abstract:
Managerial empathy ironically accelerates self-reference in predictions of consumer preferences.
The harder a marketing manager tried to think like the customer, the more they relied on their own preferences in their predictions. The paper’s explanation is precise: perspective-taking activates the manager’s own private consumer identity, not the customer’s. You sit at your desk, close your eyes, and try to be the customer. The “customer” you become is just you, in a different shirt.
The same paper found that empathic managers were less likely to use market research results. Trying to think like the customer made them trust their own intuition more, not less.
Suppression backfires when you are not sure what you think
The 2021 paper turned to the other standard remedy. If empathy does not work, can a marketer just ignore their own preferences when making the call?
It works, but only under specific conditions. If you have a strong, clearly held opinion about whether the new packaging should be blue or green, you can self-monitor: “I personally prefer blue. Is that affecting my prediction?” The answer is interrogable. The bias becomes manageable.
For weakly held preferences, the same approach backfires. The marketer asks themselves the same question. They cannot clearly identify their preference in the first place. They answer “no, my preference is not affecting my prediction,” and feel like they have addressed the bias. They have not. The preference is still doing its quiet work, just below the threshold of self-awareness.
The marketers most likely to feel they have controlled for their bias are often the ones who have done so least.
Market research becomes a mirror
The same 2021 studies showed that marketers do not consume research in a neutral way. When market research data agreed with their personal preferences, they cited it heavily. When it disagreed, they discounted it. Data was treated as confirmation, not investigation. The researchers describe this as marketers using consumer data “in an egocentric way.”
This is the trap inside the trap. Marketers who feel they have outsourced their bias to the data are often the ones whose bias is simply reflected back to them in chart form.
What this means for marketing practice
An honest reading of the research is uncomfortable and more nuanced than a single blanket conclusion would suggest.
Three of the most common tools marketers use to combat bias (empathy, suppression, and market research) can each reinforce the bias when applied without care. Empathy backfires by default. Suppression works for strong preferences but backfires for weak ones. Market research becomes a mirror unless the marketer has already done the work of separating their preferences from their predictions, which is the hard part the suppression remedy was supposed to solve.
The research authors are explicit that the bias can be corrected: when managers are made aware of the egocentric effect and explicitly instructed to suppress their consumer identity, the effect diminishes. So this is not a story of marketers being structurally broken. It is a story of standard tools that work narrowly and fail broadly.
The honest implication: the bias can be corrected from inside the marketer’s head, but only under conditions most marketing decisions never actually meet. Where it cannot, the correction has to come from outside the head.
Why NeuroBrandLab
NeuroBrandLab was built to solve the case where internal correction does not reach, which is the case in most cases.
Most marketing methods ask the marketer to predict what the customer will think, and then check that prediction against their own better judgment, more empathy, or cleaner data. All three end up filtered through the same brain that generated the prediction in the first place. The bias has nowhere to escape to.
NeuroBrandLab Signal does the measurement differently. Signal scores the message itself, not the marketer’s expectations of how the audience will receive it. Four behavioural dimensions, each derived from how brains actually decode language: Clarity Curve, Neuroscience Alignment, Perception Gap, and Mental Model Alignment. The score reflects the message’s structural decoding properties. The marketer’s interpretation of the audience never enters the calculation.
The result is a measurement that bypasses the bias documented in the research. It does not require the marketer to fix what cannot reliably be fixed from inside their own head.
“We know our customer” remains the most dangerous phrase in marketing. Not because the people saying it are wrong. Because they cannot reliably tell when they are right.
Research cited
Hattula, J. D., Herzog, W., Dahl, D. W., & Reinecke, S. (2015). Managerial Empathy Facilitates Egocentric Predictions of Consumer Preferences. Journal of Marketing Research. https://doi.org/10.1509/jmr.13.0296
Herzog, W., Hattula, J. D., & Dahl, D. W. (2021). Marketers Project Their Personal Preferences onto Consumers: Overcoming the Threat of Egocentric Decision Making. Journal of Marketing Research, 58(3), 456–475. https://doi.org/10.1177/0022243721998378
Baucum, M., & Yoo, K. (2022). Overcoming Personal Bias in Marketing Management. American Marketing Association, JMR Scholarly Insights. https://www.ama.org/2022/03/29/i-before-they-how-marketing-managers-can-avoid-personal-biases/
Ross, L., Greene, D., & House, P. (1977). The “False Consensus Effect”: An Egocentric Bias in Social Perception and Attribution Processes. Journal of Experimental Social Psychology, 13(3), 279–301.
