Representativeness heuristic

People make hundreds of probability judgments every day. Is this person trustworthy? Will this product work? Does this offer meet my needs? The intuitive assumption is that we rationally weigh all available information. In practice, however, people rely on mental shortcuts—they assess whether something matches their mental prototype rather than calculating probabilities. The question is: How strongly does similarity dominate our judgment, what errors result from this reliance—and what does the evidence tell us?

Studies

The Linda Problem

In 1983, Amos Tversky and Daniel Kahneman presented 142 Stanford University students with a personality description: "Linda is 31 years old, single, outspoken, and very intelligent. She majored in philosophy. As a student, she was deeply concerned with issues of discrimination and social justice, and participated in anti-nuclear demonstrations." The students were then asked to rank eight statements by probability, including: (1) "Linda is a bank teller" and (2) "Linda is a bank teller and active in the feminist movement." The surprising result: 85% of respondents rated option 2 as more probable than option 1—a logical impossibility, since the conjunction of two events can never be more probable than either event alone. Linda's description matched the prototype of a feminist so perfectly that participants abandoned statistical logic entirely.

The Engineer-Lawyer Experiment

In 1973, Daniel Kahneman and Amos Tversky conducted a series of experiments testing the influence of base rates. Eighty-five subjects received brief descriptions of individuals and were asked to assess whether they were engineers or lawyers. One group was told the descriptions came from a sample of 70 engineers and 30 lawyers, while the other group received the reverse ratio: 30 engineers and 70 lawyers. Participants then heard stereotypical descriptions such as: "Jack is 45, married with four children. He is conservative, cautious, and unambitious. He has no interest in political and social issues and spends his free time on hobbies like home improvement and mathematical puzzles." The surprising discovery: the different base rates had virtually no influence on participants' judgments. Both groups rated Jack as having over a 90% probability of being an engineer—based solely on the stereotypical description.

Prototypes and Categorization

Eleanor Rosch demonstrated in 1975 that categories are represented not by definitions but by prototypes. A robin is recognized as a 'bird' more quickly than a penguin, even though both are biologically birds. The reason: the robin is closer to our mental prototype of a bird. The closer an object is to the prototype, the faster and more certain the categorization.

Prototypes in Product Design

Applied to product categories: Products that closely resemble the category prototype receive higher quality ratings. A 'typical-looking' sports car is perceived as superior, even when an atypical model objectively performs better. Prototypicality builds trust and simplifies categorization.

The Art Pictures Experiment

Douglas Medin and Marguerite Schaffer conducted an experiment on category formation at the University of Illinois in 1978. Sixty-four students learned to assign abstract geometric patterns to two fictitious artists. Each 'artist' had a prototype with five characteristic features—such as specific shapes, colors, and arrangements. However, participants never saw the prototypes themselves, only variations containing 3-4 of the typical features. In the test phase, they were asked to categorize new patterns. The surprising result: categorization did not follow the abstract prototype but rather the similarity to the concrete training examples they had seen. Patterns that resembled many training examples were categorized faster and more confidently—even when they were further from the theoretical prototype. The observed exemplars, not the abstract rule, determined the category boundaries.

The Law Students Study

In 1977, Lee Ross and colleagues at Stanford University investigated how law students work with legal precedents. They presented 90 students with two court cases involving a fictitious legal problem. Half the students received two very similar cases, while the other half received two different cases. The students were then asked to evaluate a new case. Those who had studied similar example cases applied the underlying principles much more narrowly—only when the new case strongly resembled their examples. In contrast, the group exposed to different examples recognized the broader principle and applied it more flexibly. A single example, or very similar examples, narrowed students' conception of when a legal principle applies. Only exposure to a range of different examples enabled transfer to new situations. The concrete examples shaped students' mental category boundaries more powerfully than the abstract legal rule itself.

Principle

Which principle for Customer Experience Design can be derived from this? The Representativeness Heuristic shows that customers evaluate products and services primarily based on how closely they resemble familiar prototypes, rather than on objective facts or statistics. For successful customer experience design, brands should deliberately design their offerings to align with the mental categories and expectations of their target audience. This approach is particularly effective for complex or new products where customers need guidance—however, overemphasizing similarity can create confusion with highly differentiated offerings. The following guidelines demonstrate how to implement this principle in practice.

Guidelines

Show prototypical example customers

Rather than presenting abstract target audience descriptions, showcase concrete example customers with names, photos, and stories. Potential customers instinctively ask themselves: 'Am I like this person?' This sense of perceived similarity serves as a more powerful purchase trigger than demographic data alone. Crucially, the examples must be sufficiently diverse to allow different customer types to identify with them.

Activate known categories

For innovative products, explicitly communicate similarity to familiar categories: "Like Spotify, but for audiobooks" or "The Tesla of e-bikes." These comparisons activate mental prototypes and facilitate understanding. Without this bridge, customers perceive new offerings as risky because no comparison category exists. The key: Choose the right reference point that evokes positive associations.

Using visual prototypes

# Improved Text In visual design, deliberately activate prototypes: Medical apps use white and blue because these colors align with the prototype of 'trustworthy medicine.' Financial apps rely on dark tones and crisp typography that signal credibility. These visual codes work because they match users' mental categories. Designs that break these expectations must justify the additional cognitive effort they demand. --- **Changes made:** - "consciously" → "deliberately" (more precise in design context) - "correspond to" → "align with" (clearer, more direct) - "clear typography" → "crisp typography" (more specific design terminology) - "the mental category" → "users' mental categories" (clarifies whose mental categories and uses plural for accuracy) - "Those who break" → "Designs that break" (clearer subject—designs, not people) - Added "they demand" (completes the thought more clearly)

Describe typical use cases

Instead of presenting feature lists, describe prototypical use cases: "At 7 AM, still in bed, you quickly check the most important metrics." Customers unconsciously ask themselves: "Does this fit my daily routine?" The greater the similarity between the described scenario and their own life, the higher the perceived relevance. Important: The scenarios must be authentic and reflect real customer situations, not marketing fantasies.

Respect category conventions

Respect category conventions Familiarize yourself with the established patterns of your product category. A sports car should look like a sports car, a banking app like a banking app. Innovate in the details, not in the basic form. The following examples illustrate this guideline:

• Automobile: Tesla is innovative in powertrain and software – but a Model S looks like a sedan because customers expect it to.

Introduce innovation gradually

Introduce radically new elements gradually, embedded within familiar structures. Too much novelty at once overwhelms and creates rejection. The following examples illustrate this guideline:

• Apple Watch: The first Apple Watch looked like a watch – familiar form, new technology. If it had looked like a wrist computer, acceptance would have been lower.

Familiar Elements as Anchors

Use familiar elements as 'anchors' to contextualize new information. 'It's like X, but with Y' helps with categorization. The following examples illustrate this guideline:

• Slack: 'It's like email, but for teams in real-time.' The email anchor helped people understand a new category.

Demonstrate diversity in use cases

Present deliberately diverse examples of product usage and customer profiles. Avoid showcasing only the ideal case or a homogeneous user group. A SaaS tool shouldn't feature only tech startups—include SMEs, non-profits, and sole proprietors as well. An insurance company shouldn't show only young families—represent singles, seniors, and self-employed individuals too. The range of examples determines who feels addressed and what seems possible. The more diverse the examples, the broader the mental category.

Omit examples of non-target audiences

Negative examples ("Not suitable for...") can unintentionally narrow category perception. Explicitly stating "Our software is not for small teams under 10 people" makes this boundary more salient than necessary. Better approach: Show positive examples of your target audience without mentioning non-target groups. Exception: When the distinction is purchase-critical, such as technical requirements. In these cases, frame it as a factual requirement ("Requires SQL knowledge") rather than a negative example ("Not for SQL beginners").

Initial contact with representative specimen

The first example shown becomes the mental anchor for the entire category. In product demos or website content, the initial example should be deliberately chosen: Does it demonstrate the core function? Does it appeal to the primary target audience? Is it neither too simple nor too complex? A CRM tool that first shows an enterprise setup with 50 custom fields will lose small and medium-sized business customers. One that starts with a two-person team will deter corporations. The entry example should match the median complexity of the target audience.

Leading from the concrete to the abstract

Begin with concrete examples, then transition to the abstract principle. "Customer A uses it this way, Customer B uses it differently—the underlying principle is..." This sequence harnesses the strength of examples (comprehensibility) while avoiding their weakness (overly narrow interpretation). Relying solely on example-based communication forces customers to extract the principle themselves, which may lead them to derive the wrong rule. Explicitly stating the abstraction after presenting multiple examples corrects their mental model and enables them to transfer the concept to new situations.