Minor mistakes may be doing something unexpected in customer service: making interactions feel more human. A new study co-authored by Associate Professor Juliana Schroeder suggests that slight imperfections can soften how people judge automated and human agents. The findings arrive as companies race to add AI into frontline service, raising new questions about trust, tone, and what customers actually prefer.
The research points to a simple idea with big consequences: people respond not only to accuracy, but also to warmth. When tools get every detail right, they can feel cold. When they slip a little, they can seem more like us.
Background: The Appeal of Imperfection
Psychologists have long studied how mistakes shape likability. In the 1960s, the “pratfall effect” showed that a high performer who blunders can seem more relatable. The insight has stuck. Brands, leaders, and creators often weigh whether to show polish or personality.
Now, that debate is moving into AI. As customer service shifts to chatbots and voice assistants, companies must decide how perfect these systems should sound. Errors can frustrate. But flawless scripts can feel stiff. The study by Schroeder and colleagues tests that trade-off in modern service settings.
“To err is human…and in the age of AI, it may be humanizing. A study co-authored by Associate Professor Juliana Schroeder found that people view customer”
What the Research Suggests
The work indicates that small, harmless errors can increase perceptions of warmth and humanness. That effect appears especially relevant when people suspect they are dealing with an automated system. A slight slip can signal there is a person—or a person-like mind—on the other side.
But there is a catch. Warmth can rise as competence seems to fall. Customers may like an agent more after a tiny mistake, yet trust it less for complex tasks. The balance matters most in high-stakes issues like billing, health, or safety. There, precision still wins.
- Minor errors can boost perceived warmth and humanness.
- Competence judgments may dip when errors occur.
- Context matters: low-stakes chats differ from high-stakes requests.
Customer Experience and Design Choices
For service leaders, the message is not to make systems sloppy. It is to design for tone as well as accuracy. A brief apology, a natural pause, or a simple acknowledgment can soften an exchange without adding real mistakes.
Some teams already test small style changes, like contractions, casual greetings, or limited self-correction. These touches can create the feeling of a conversation, not a script. The study’s framing suggests that warmth cues matter, even when the agent is an AI.
Risks and Boundaries
Imposed errors are risky. Customers do not want staged typos when their card is declined. In regulated sectors, any mistake could trigger real harm. The safer path is to add human signals without adding wrong answers.
There are also fairness and transparency concerns. If people read warmth as “more human,” they may over-trust an automated agent. Clear disclosures about AI use can help set expectations and protect users.
Industry Response and Next Steps
Banks, airlines, and retailers are watching how tone changes outcomes like satisfaction, repeat use, and escalations. Many are moving to hybrid models where AI handles routine items, and humans step in for judgment calls. The study supports that split: let machines be precise on simple tasks, and let people lead when stakes rise.
Experts recommend careful testing. Track error costs and customer sentiment together. Measure if small warmth cues ease frustration during wait times or policy explanations. Do not assume one style works for every channel or audience.
The new research adds a timely reminder: performance is not only about getting it right. It is also about how it feels. As companies roll out smarter tools, they should build systems that are accurate, clear, and approachable. The next phase of customer service will likely reward both precision and humanity. Watch for more trials that blend the two—and for clearer rules on when a gentle touch helps, and when only an exact answer will do.
