It’s no novel concept anymore that chatbot surveys are used for market research as an alternative to traditional surveys. Even though the practice of chatbot-led research conversations is still in its cradle, there are bold statements indicating that chatbots are the future of market research. 

A promising future aside, we have heard plenty of questions and skepticism from our research and insight community—why do we need a chatbot for market research? Is it really better, or is it just another fad? 

We build an AI-powered insight platform centered around chatbot technologies. From own experiences, we feel that there are a couple of important concepts that are worth clarifying to help paint an objective picture of market research and feedback chatbots. Two considerations most cited as reasons why chatbots are advantageous for consumer insights than traditional surveys are as follows:

1. Chatbots are messaging apps

For social interactions, people love messaging apps; it has an interface that feels natural and comfortable to many, so much that some people practically live in those apps. People’s love of messaging apps for social interactions, however, doesn’t automatically make them love chat-based market research. When people chat with their friends, they want to chat. The chat format is not a driver, but an enabler or a facilitator, at the best. Surveys masked in a chat format aren't going to make people want to take surveys more.  

2. Chatbots for research = chatbots for customer service

More and more chatbots are being used for customer service. Gartner predicts that chatbots will power 85 percent of all customer service interactions by the year 2020. The increased popularity of chatbots used for customer service, however, doesn’t prove the same concept for chatbots for market research. Yes, a chatbot is a question-answering system that leverages Natural Language Processing (NLP), but customer service bots are trained to answer questions while market research chatbots are, supposedly, trained to ask questions. The latter is a much harder task and expectation to manage.

Having said that, chats do have some natural advantages over forms, the format that the vast majority of surveys use. Chats are associated with a natural and often fun part of everyday life. It’s supposed to be conversational and effortless. Traditional survey platforms also find chats appealing and have been offering chat formats instead of just forms. However, this approach is just an inauthentic placeholder that does a great disservice to real chatbots.

As shown in this example, the classic survey format (form) and the conversational chat format appear to be easily interchangeable on some DIY market research platforms.

Imagine having a typical survey question such as “next you will see a list of statements, please indicate your level of agreement with each using a 7-point scale…” and shoving that into a chat bubble… will that make it a “conversation”?  

The difference between a “classic” form-based survey and a “conversation” is not simply just the look.   Conversational research means much more than an interface make-over. 

If chatbots have the potential to come to the rescue of surveys, we need to lead genuine conversations to bring out its full potential. Here’s our take on how to do that: 

1. Good listening and show it! Chatbots create the impression that someone else on the side is listening. Respondents say it motivates participation if they feel they are heard. Appropriate acknowledgment and segues throughout the conversations are effective ways to show good listening. 

2. Ask smart follow-up questions. Asking intelligent questions is not just for the sake of the richness of research findings but also an effective engagement measure. Probing for deeper insights shows that bot understands and cares. It also conveys the importance that the participants should think harder, too. 

3. Build rapport: Chat scripting is what we call questionnaire design. If we ask the right questions, using the right language and the right tones for the right audience, then we know they will reciprocate. We know that the extra effort put into scripting pays off when we receive vivid consumer language in return. 

4. Interface matters! We need to create an interface to support both open-ended and close-ended questions to optimize the chat experience and research outcomes, but, of course, the close-ended questions need to be designed to fit the chat. There’s an ocean of opportunities to create smart “survey” questions. That definitely deserves a separate blog post. 

5. Come back to the basics. We often need to remind ourselves why we need a chatbot in the first place. To us, chatbots allow one-on-one attention, regardless of the number of people we are chatting with. One-on-one attention is what’s lacking in traditional survey research and is fundamental to engagement and data quality. We’ve done a lot of analysis and reflection on that and will share our thoughts in a separate blog post, but that guideline always helps us stay the course.

So, are chatbots the silver bullet for market research? Our answer is it depends on how chatbots are designed and used. We are at an early stage of mastering research conversations. We would love to hear your thoughts and perspectives. If you also find this topic exciting, we invite you to sign up for our monthly newsletter, “the Nexxt Iteration”, for the latest development and our fresh perspectives as we continue charting this new territory.