Accompanying beginners

Trust demands realistic self-assessment—of one's competence and knowledge. Yet some people overestimate themselves while others underestimate themselves. The question is: Is there a systematic relationship between actual competence and self-assessment, and how does this influence customer relationships? What evidence exists on this topic?

Studies

Unskilled and Unaware of It

Justin Kruger and David Dunning conducted a groundbreaking experiment at Cornell University in 1999 that revealed a fundamental cognitive bias. They tested 65 students in logic, grammar, and humor, then asked them to estimate their own performance relative to others. The bottom 25% of participants, despite scoring at the 12th percentile, rated themselves at the 62nd percentile—a dramatic overestimation of 50 percentage points. The top performers, conversely, slightly underestimated themselves. The striking finding: those who needed help most were least aware of their weaknesses.

Training improves self-assessment

In a clever follow-up study, Kruger and Dunning tested whether competence improves self-perception. They provided the weakest participants with a 10-minute logic training session, then asked them to re-evaluate their original test performance. The results were striking: their self-assessment dropped from the 61st percentile to a more realistic 44th percentile—despite evaluating the same performance. The training had sharpened their metacognition. Only by understanding sound logic could they recognize their own errors. Paradoxically, competence enables the recognition of one's own incompetence.

Why Experts Underestimate Themselves

The top performers in Kruger and Dunning's study demonstrated the opposite problem: they estimated their performance at the 70th percentile when they actually ranked in the 86th percentile. When researchers showed them other participants' actual answers, the experts revised their self-assessment upward to the 82nd percentile. This underestimation stemmed from a false-consensus effect: they assumed their exceptional abilities were normal. Nevertheless, their self-perception remained significantly more accurate than that of lower-performing participants. In essence, experts underestimate how rare their competence truly is.

Principle

Which principle for Customer Experience Design can be derived from this? The Dunning-Kruger Effect reveals that novices systematically overestimate their ability to use products and make informed decisions. This overconfidence arises paradoxically from their lack of knowledge—those who don't understand a system's complexity cannot assess what they don't know. For customer experience design, this means companies must proactively provide support and guidance, even when users don't explicitly request it. This effect becomes particularly critical with complex products or important purchase decisions, where overconfidence can lead to frustration and poor choices. The following guidelines demonstrate how to implement this principle in practice.

Guidelines

Make progress continuously visible

For longer processes or content, continuously display progress using scrollbars with chapter markers, "You are at step 3 of 7" notifications, or estimated time remaining for forms. The mechanism: progress indicators create small goals and moments of achievement that sustain attention. Without orientation, the process becomes a monotonous, endless loop—ideal conditions for mind-wandering. Particularly effective: non-linear progress bars that fill faster toward the end, leveraging the goal-gradient effect.

Proactively communicate wait times

Uncertainty intensifies frustration—proactively communicate wait times using progress bars, estimated completion times, and regular updates. Providing information gives users a sense of control and makes time feel like it passes more quickly. The following examples illustrate this guideline:

• Domino's Pizza Tracker: Real-time tracking: 'Your pizza is being prepared', 'Your pizza is in the oven', 'Your pizza is on its way'. The waiting time feels productive because something visibly is happening.
• Disney FastPass: Displayed wait time is systematically overestimated. '45 minutes' shown, but served after 35 minutes. Positive surprise instead of frustration about the wait time.

Reveal options step by step

Complex products should begin with simple defaults and progressively reveal advanced features. Conceal additional options behind 'Show more' links to prevent overwhelming beginners, while ensuring power users can access them when needed. Each new level signals to users: there's more to discover when you're ready. The following examples illustrate this guideline:

• Google Suche: The homepage shows only a search field. 'Advanced search' is available but hidden. 95% of users don't need it.
• Slack: New users see a simple interface. Advanced features (threads, reactions, apps) are introduced gradually – when the user is ready.