Most Chatbots Still aren’t Smart Enough: Survey


As per the survey, 72% of respondents generally find these automated bots to be helpful to an extent, but the quality of interaction can be relatively mixed

With rising use of virtual customer assistant (VCA) or chatbot technology by companies in customer support operations, majority of people still think that most of the chatbots aren’t smart enough to fulfil the high expectations. 65% of people still prefer a human agent at the other end of the chat, according to a survey conducted by Pegasystems Inc.

Whereas most of the respondents agree that chatbots can be convenient and fast in certain situations, they mention lack of intelligence as their major complaint against chatbots, as per the survey.

According to the survey, 72% of respondents generally find these automated bots to be helpful to an extent, but the quality of interaction can be relatively mixed. 58% ranked their experiences with chatbot as just ‘adequate’, i.e., doing several tasks well and others not properly.

Around 18% of consumers complaint that automated bots are ineffective and even annoying. Just 16% respondents gave a high quality rating for their experience with chatbot technology, the survey noted.

Chatbots are preferred for only simple queries

As per the survey, customers prefer chatbots only for simplest queries and cases that can be done rapidly – tracking order (60%), finding basic information (53%), and asking basic questions (49%).

According to the respondents, the top benefits of chatbots are fast service (56%), ability to engage on their own schedule (37%), and convenience (36%). If done well, 43% consumers believe that chatbots can be nearly as good as interacting with human, whereas 34% does not agree, and 23% are still not clear.

Lack of human touch

Most organisations claim that artificial intelligence (AI) power their chatbots. However, consumers, based on their experience, still have complaints about the bots. 27% consumers think that bots are not enough smarts to effectively answer their questions, 24% complaint about lack of context in the conversation, and 14% consumers complained about robot-like engagement with few human qualities.

Similarly, the major reasons when the consumers would drop a session with an automated bot include when bots cannot answer their questions (47%), make them do more work than expected (47%) or are too vague in how they can assist them (43%). On the other hand, just 17% consumers responded that they would use a chatbot for purchasing goods and services, creating uncertainty on the objective of linking bots to direct revenue.

Non-users are still reluctant

The survey is also conducted on a separate group of customers who have not used the bots yet. It found out that the reasons for not using a bot include lack of any real exposure to chatbots (53%), personal preference to engage with human only (30) and lack of knowledge on how to use chatbots (23%).

Around 45% of such consumers, having no chatbot experience, responded that they’re even not planning to experience the one in coming year whereas 30% are not sure about to use in near future. This presents a hurdle to organisations trying to enhance reliance on digital channels.

According to the survey, 25% of the non-users said they are willing to experiment with a chatbot, but with some reluctance. The major concerns arise from their lack of experience about chatbots. 46% consumers simply don’t know how to use chatbots, 31% respondents think they lack confidence in chatbot effectiveness and 27% have worries regarding security and privacy.

Pegasystems conducted the survey on 3,500 consumers across different geographies including the US, the UK, Canada, Australia, France and Germany. The survey is conducted on two different groups – consumers having chatbot experience and consumers who have not used chatbots.


Please enter your comment!
Please enter your name here