My Unlikely Road to Tech Entrepreneurship

Originally given as a talk at the MRIA Annual Storytelling Event in Toronto in the summer of 2017

Good evening, everyone. My name is Kathy Cheng and I am the founder of a technology start-up, Nexxt Intelligence Inc. We’re building an artificial intelligence-powered research platform to better understand consumers, citizens, and employees. We focus on creating experiences that are more meaningful, delightful and, ultimately, more human for our research participants. When participants are engaged, their reactions are genuine, and insight flourishes and innovation happens as a result.

Today I’m going to tell you about my unlikely road to tech entrepreneurship in market research. Why unlikely?  I’ll tell you what happened to me 12 years ago on my first day at my new job at Ipsos. I sat down with our IT guy, who was helping me set everything up on my laptop. I was basically following his instructions step-by-step and he was being very patient with me. When he asked me to type something like “Ipsos” for example, “followed by a backslash,” I did exactly what he said, but instead of typing the symbol “\”, I typed “backslash” letter by letter. He was completely stunned and amused, and I was completely embarrassed. He said, with all good intentions, “I have never come across anyone doing this.” I concluded on the spot that my technology quotient was close to zero and I would not want to have anything to do with technology in the future.

But 10 years later, how did I get myself into technology? Despite the fact that it was obviously a very deliberate and conscious decision, I have to believe it was a less conscious thought process that fundamentally drove me and continues to drive me today, especially when the reality of having to learn and deal with technology kicks in. Those are the powerful imprints during the early days of my career that molded me into who I am today.

I started my career in market research back in Shanghai about 20 years ago; first as a simultaneous interpreter for qualitative research. Back then, the vast majority of market research was commissioned by international brands trying to establish or grow their presence in China. Clients were mostly English-speaking and research was conducted in Chinese languages/dialects. It was a perfect job for an English major student, so I did it for a couple of years. Upon graduation, I was offered a job by Nielsen to become a qualitative researcher. I did not realize the huge value of the time I spent as an interpreter until I became a moderator myself. The opportunities to have observed moderation from top moderators as an outsider allowed enough distance for me to understand the dynamics between the client, moderator, and respondent, and provided a more holistic perspective. My boss, and my mentor, to whom I feel forever obliged, taught me many things that are still very true 20 years later. For example, she said a successful group needs three things: First the client is happy that they feel they’ve learned all what they wanted to learn from the group; second, the moderator is happy that he/she asked everything he/she wants to ask; and last, but not least, that the respondents are happy that they feel they’ve shared everything they wanted to share.

I thought the last part was particularly important and true; that’s the necessary condition for the first two. And I’ve carried her advice with me throughout my career for the past 20 years. The greatest satisfaction I have is always when, at the end of a group, respondents tell me that they enjoyed the discussion, or that they felt they learned something from others, or that they felt they had a good time sharing their thoughts. Or that they gave me their contact info and wanted to be invited to more discussions as such (not that I am allowed to), or in other words, in our language, to see how engaged they were. A few years into my career, I started to do more quantitative research; respondent engagement is equally crucial here, although it’s not as easily observable during the process of data collection, but definitely observable in results.

The role of the “respondent” is, however, significantly biased by the very term “respondent.” I remember my mentor telling me after she came back from an ESOMAR conference how inspired she was by a talk about retiring the term “respondents,” because “respondents” suggests more of a passive role, while in reality, they are our key partner, an active participant in research. Again, that was almost 20 years ago, but interestingly enough the term is still widely used – me included, just for convenience, despite my clear dislike of its semantics; and I am sure many of you would agree with me.

Now that respondent engagement has become increasingly challenging, a hot topic of the industry, I am constantly reminded of the advice my mentor gave me. That I started my career as a qual specialist is useful. To me, “respondent” engagement is a natural requirement, a given, that determines the quality of the research.

Until recently, we researchers have to face an increasingly challenging reality: What if we can’t rely on our respondents to respond to us anymore? What if it’s harder and harder to engage them simply through monetary incentives? What if respondent engagement is the Holy Grail for true insight?

Those kinds of questions puzzle me, bug me, make me feel uneasy and make me want to find a solution – especially when our job is often to advise our clients on how to be customer-centric and how to employ design thinking to better engage their customers.

What if we, as a research industry, tried to be “respondent”-centric? What if we actually turn “respondents” into “participants”?

We qualitative researchers, especially ethnographers, are natural design thinkers. The whole point of conducting ethnographic research is to live their lives, to walk in their shoes, isn’t it? How about we apply some of that way of thinking to research in general?

Instead of us making respondents answer the questions we want them to answer, maybe we should try to hear what they have to say. Instead of ushering respondents to a made-up research platform, we should meet them where they are, whether it’s social media, at home or at work.

So how to become “respondent”-centric was a question I had been thinking about for a while, until one day I felt I had this moment of enlightenment watching my 12-year-old having this endless conversation with Siri:

“What’s your name, Siri?”

“My name? It’s Siri!”

“Siri, how old are you?”

“I am old enough to be your assistant.”

“Siri, what’s 0 divided by 0.”

“Imagine you have 0 cookies and you split them evenly among 0 friends, how many cookies does each person get? See, it doesn’t make sense. And Cookie Monster is sad because there are no cookies, and you are sad because you have no friends.”

He was having so much fun. He was totally engaged. And he was not paid to do that.

Imagine if our “respondents” participate in our research because they want to! Technology can be a solution, or probably it has to be, especially for the younger lifelong digital natives.

I was reminded of what Steve Jobs said, “It’s technology married with liberal arts, married with humanities, that yields us the result that makes our heart sing.” The how-to piece those elements together took us much longer to brainstorm, or prototype, to try and fail, but we’ve been very clear about what we want to achieve.

We want to make our respondents’ hearts sing.

If that’s what I wanted to do, there was also a clear need for a big leap of faith on my part. I mentioned earlier that I believe my decision to start up a tech company was driven more by imprints – the more subconscious desires for my career. That was not a random comment. I said that based on a lot of research and actual experience. Cultural research has always been an important part of my career. Culture as an imprint has its unconscious influence which affects our thoughts, emotions, behaviours, and habits. When it comes to understanding multicultural consumers, just to hear what people have to tell us is often not enough. It’s why they say so that is more meaningful. I’ll give you an example. I recently tested an advertisement in focus groups, the ad shows a girl wearing a hot pink dress as the winner, while the other girl wearing a beige blouse and skirt is a loser because she was too slow in making her purchase decision. The message is if you want to be pretty and get what you want, you need to buy immediately; otherwise, you’ll miss out on your opportunity. The message seems to be clear and resonates well with mainstream Canadians, but my Chinese respondents were all looking puzzled after seeing the ad a couple of times. They found the ad hard to understand and all they could say was “but the ‘loser’ girl looks prettier.” What’s driving their reaction to the ad is the cultural DNA. Chinese are from what’s called a more “neutral” culture vs. “affective;” they naturally gravitate towards subtler hues, and find showy things like the girl in the hot pink dress less attractive.

It became increasingly clear to me that we needed a solution that goes beyond asking people – especially through traditional questions such as, “On a scale of 1-10, to what extent do you agree that this describes you?” – because if it’s the unconscious that we are trying to understand, we can’t expect people to tell us that, can we? There’s only so much we can learn from what people are able to articulate. Like Henry Ford said long ago, “If I had asked people what they wanted, they would have said faster horses.”

We researchers need to connect the dots between what people can articulate, and what can be modeled and predicted through the use of technology.

Yesterday, my very fun colleague suggested that I dress up as a robot for today’s event to communicate that we are at the cusp of new cutting-edge technology and methodological breakthroughs. I was amused by the idea and appreciated the creative thinking, but I had other meetings to go to before this, so I said to her “The use of technology is to be more human, so I will dress like a human.” With that, I will conclude my story for tonight. Thank you all very much for listening to what I have to say, and I welcome any questions or thoughts that you may have for me.