Wie man dem Uncanny Valley ausweichen kann:
Menschlich sein (echt): Will man eine Marke durch Automatisierung vermenschlichen ist oftmals nicht der richtige Weg. Viele Schritte können leicht von einem Programm übernommen werden. Der beste Botschafter einer persönlichen Nachricht ist nun mal der Mensch.
Die Kundenbedürfnisse sind wichtig: Man liebt eine Banking-App oder Cloud Storage weil er funktioniert, nicht weil es dir Komplimente macht. Gute, userzentrierte Gestaltung erzielt deshalb oft ein viel positiveres Gefühl als Emotive Scripts.
Maschinen sind Maschinen: Der User steht auch hier immer im Zentrum. Deshalb ist es wichtig sich auf seine Erwartungen/Erfahrungen zu konzentrieren und keine unangenehmen Situationen zu erzeugen. Deshalb reicht es nicht einem User zu erzählen wie gut er kochen kann wenn das vielleicht garnicht Teil seines Interessensbereichs ist.
Transparenz: My internet provider always signs its emails with: “Kind regards, Telfort Customer Service“. They do address me with my name. Likewise, some weeks ago I couldn’t start a support chat with my bank ING without first providing my name. But the support rep I was connected to didn’t return the favor; it just said “ING service”. These are clear cases from the uncanny valley of customer service. An information asymmetry that’s alien to any normal human interaction. Whether it’s in writing or voice, a name should be part of any service interaction. Similarly, it’s weird to have a chat when the other side’s profile image isn’t the picture of the person.
Legeres/bodenständiges Vokabular: Artikel
As brands adopt more automation in their social media, bots, IVR systems, marketing programs, and customer care systems, they must be careful that the desire to seem more human doesn’t inadvertently cause negative, brand-damaging experiences.
Chatbots sollen Chatbots bleiben: Chatbots are receiving plenty of media attention and are expected to play a major role in customer service. One company that’s been using a chatbot for support for quite some years is IKEA. If you ever visited their website, you might be familiar with chatbot Anna. Anna has recently been taken to the robot cemetery, but she’s answered online questions for over 10 years.
Anna is, however, a perfect specimen from the uncanny valley of customer service. Has she been successful? I guess so. But I’d argue that she’s been successful despite, not due to, her configuration.
The problem with Anna was that she was neither clearly robot or human. Her name, image, and texts were ambiguous. When you chatted with her for the first time, you might have thought you were chatting with a human – or at least with an intelligent online assistant.
But as David Meerman Scott pointed out, her limited answers were disappointing, and she was actually nothing more than an online greeter. She was neither human nor robot, the chatbot equivalent of Japanese humanoids.
Although Slackbot can be useful, he’s very clear about his lack of intelligence. He sets the right expectations, and he doesn’t freak us out. Because of that, Slackbot is on the right side of the uncanny valley.
Emotion isn’t a tactic in customer experience; it is an outcome.
