April 9, 2024
Welcome to the second installment of our multi-part series delving into the influence of artificial intelligence (AI) in the user experience (UX) industry. For this discussion, we had the pleasure of interviewing Nitzan Shaer, the CEO and co-founder of WEVO.
WEVO is a key partner of CXperts, offering a unique UX research solution that leverages AI to deliver reliable insights in a fraction of the time traditional tools require. With 100+ panelists in a WEVO Pro study, companies can quickly amass qualitative and quantitative insights at scale.
Nitzan has continuously been fascinated by what drives people to make decisions and, under his leadership, WEVO has been able to deliver reliable user insights and better experiences that designers, marketers, and product owners can rely on.
We are excited to share Nitzan’s valuable perspectives on the dynamics between AI and UX and how WEVO has become an innovator in the space.
Nitzan:
I always try to look at it from the perspective of what problem can AI potentially address.
I think one of the challenges facing user researchers today is time, they do not have enough time to conduct all the user research that they would like to do. And this is true for user researchers, for designers, for marketers, for product managers. They have a lot of things they need to do. We are in the age of efficiency.
And hence all the people that consume user research - marketers, user researchers, designers, and product managers - need to find ways to do more with less. So I think to that end, one of the key things we'll see AI doing is helping them save time by doing the heavy lifting. In many cases, multiple elements of the user research can be done by AI and we will find more and more solutions where AI is doing that.
Nitzan:
As a smart man once said, “I don't believe AI is going to take your job, but a person using AI may take your job,” which is to say we're strong believers at WEVO about augmenting the work that user researchers do with the help that AI can provide. I think that combination - that augmentation - is where we're going to see the market go to.
For instance, the work of summarizing key findings from user research and pulling out themes from multiple comments that users have made. Another use is the ability to have a conversation and to ask questions about a data set that I’ve just collected, like a chat with the AI helping to address the questions. Those are already some of the early uses of AI that we are seeing. This is all in the service of doing everything faster and, ideally, in a more reliable way as well.
Nitzan:
There are four or five ways that our product is leveraging AI currently:
1. Key Findings
Key findings are generated for the results of our 120 person studies. With regular tools without AI, it would be very time consuming to summarize data from 120 people, but with AI, we enable our customers to get more reliable data compared to 8-10 person traditional studies. The AI does the heavy lifting, putting together the first draft of those insights, but we do have a human researcher again, augmenting the AI, going through the findings to ensure smart, accurate output. So it's a partnership there between a human and an AI.
2. High Quality of Participants
The second way we are using AI at WEVO is to ensure the high quality of the participants. It’s kind of like having an AI sitting and observing the feedback that people are giving. It is very good at assessing: are they concentrating, are they answering to the point, are they who they say they are? That AI ensures that there is a high quality of feedback that’s being received and then later down the road can be analyzed.
3. Extracting Themes
The third area is taking large volumes of data and finding the most important themes and expectations. For a WEVO study, the AI will extract the most common customer expectations and evaluate to what extent those expectations are being met based on the user feedback. Again AI is doing the heaving lifting work there.
4. Prioritizes More Quality Feedback
The AI also pushes the top quotes or pieces of feedback to the top of the page within the WEVO results. That’s again saving time for the user researcher to read through all the quotes. For instance, if you have three pages of user quotes to go through, the higher quality quotes are the ones surfaced at the top.
5. AI-Based Samples
AI based samples, or silicon samples, is another emerging space in user research that is very, very new. This is feedback that is not leveraging human participants, but rather getting feedback from the AI by saying, “hey, without any human feedback. What do you think of the strengths and weaknesses of this user experience? What about this body of text that I just fed you? How would people out of this demographic, this type of persona, respond to this content?”
AI Based samples are being used in WEVO Pulse, our newest product we recently released that is leveraging generative AI. It enables us for the first time to get feedback on designs, concepts, and innovative ideas within less than five minutes. So suddenly, I can now test and get feedback in cases where I could never do it before - if I had a deadline in an hour, or if I want to do five iterations today, or if I didn’t have budget for a big user study.
Nitzan:
We don’t recommend WEVO Pulse as a replacement for a user study. If I need to make a big decision and it’s a multimillion dollar campaign or a million dollar product, I still want the feedback from 120 people through WEVO Pro. But for all those small iterations along the way, that I could never afford to do user research on - not from a time or a budget perspective - I could do that with AI and that's a really interesting opportunity that's opening up now if you do it correctly.
There are some people that come to us and say, “What do you mean you're doing user research without users? That's preposterous. That's blasphemy.” And I respect that. But I think as with all new technologies, we should remain open to evaluate its strengths and weaknesses. Is it as good as user research? Absolutely not. Should it replace user research? As it is today, absolutely not. But can it be used in its infancy? It's in its infancy. Can it be used to get feedback in situations where people can't conduct user research? Is it better than nothing? It absolutely is.
Nitzan:
To them I would say test it and see for yourself. See what it can do. What are the insights you're getting from this tool? Compare that to the insights that you're getting from your typical user research? Then pass judgment and see if it’s good enough. I can tell you that is what we are doing. We've been doing that for the past five months continuously. We're comparing a WEVO Pro study with 120 people to what the AI says. That's how we're making the system better and better and better all the time again.
I’m very pleased to be able to share these highlights from my discussion with Nitzan. As a UX Researcher and daily user of WEVO, I can say with first hand experience that AI is streamlining the amount of user research we are able to conduct and increasing the quality of the insights. As Nitzan and I discussed, we still have our real users whose feedback is more valuable, more authentic than AI based samples, but we are starting to see how AI insights can start to supplement our work.
I love the perspective that we can do research on AI itself and evaluate its efficacy. As UX researchers, let’s do our own research on the results of AI insights. We can be inquisitive and curious about the potential uses of AI and see how it performs. There is a partnership between humans and AI that is going to save us time, help us implement new ideas, and allow for continued innovation.
If you would like to read more about AI’s emerging influence in other aspects of UX work, take a look at our first article, A Conversation with AB Tasty.
Finally, stay tuned for the final part of this series, where we continue to share illuminating ideas from our partners at Fullstory.
As always, reach out to us at hello@cxperts.io and we can have some AI & UX chat’s of our own.
Resources:
https://wevo.ai/introducing-wevo-pulse/