The new insight engines that transform researchers into insight creators
In this series we speak with tech leaders who offer solutions within the research, data, and insights space. To shine a spotlight on disruptive technologies and better understand the value they bring.
In our latest instalment, we spoke to Kathryn Topp, CEO & Founder of Yabble, a platform that uses generative AI and machine learning to create instant insights from unstructured data, enabling researchers to speed up the analysis process.
At Day One, we’ve been working with Yabble, taking advantage of the benefits of GPT technology, combined with Yabble’s proprietary smarts all within a secure and compliant environment. We recently used Yabble on a fast turnaround qualitative study, to support our human analysis to identify and create patient profiles that were considered suitable for our client’s brand. With just three days to go from analysis to report, Yabble helped us to interrogate the data, hypothesise the profiles and increase the speed at which we were able to deliver the research. But now, let’s delve into the interview itself.
Kathryn, could you tell me a little bit about yourself and the genesis of Yabble?
I love research and understanding the whys behind consumer behaviour. I have always had a real thirst for knowledge, curiosity, and a love of using technology to help us be more effective and efficient. I set up one of the world’s first customer communities at a time before anyone really knew what a community was. There were so many gaps in the market, in research, that could be addressed by tech. And so, the first iteration of Yabble focused on affordable DIY software to build your own communities. We then began thinking about what some of the really big, juicy problems are that are out there in the industry such as how can we extract insight from unstructured data using generative AI. We challenged ourselves to ask, ‘is there a different way to solve this problem that is more effective for researchers?’ Ultimately, Yabble is about saving time and freeing the minds of researchers to allow them the space to apply strategy based on the insight.
How exactly does Yabble ‘free the minds of researchers’?
We have two clear problems that we solve for. The first is, how do we more easily and quickly unlock insight from unstructured data? The second is, how do we access quality data more easily to answer research questions?
And how does Yabble do that?
Yabble engines are 1000 times faster than a human at processing data. We automate data processing, cleaning and analysis. You can then interact with and code the data in lots of different ways. You can ask your data questions and continue to question it as your data grows, and as you augment it with more layers, to get ever deeper insight.
Can you give me some examples of where Yabble has helped?
Customer experience is a common area for a lot of our customers. One of our customers is collecting NPS scores and thousands of pieces of data every week. They need to take all that customer feedback and turn it into something they can utilise in their organisation at an individual store level, but also at a management level. Review data is also popular, and we provide the engines to help customers understand how brands and products are performing. As well as being a tool for traditional researchers, Yabble can also help what we call the next generation of researchers, people who might not be traditional researchers, to make sense of human behaviour.
How does the engine work?
We’ve trained our engines on over 865,000,000 pieces of unstructured data. We’ve had over 50,000 hours of training go into our engines. We use large language models, including GPT models from OpenAI. We then layer this with Yabble’s proprietary smarts so that when customer data comes in, we automatically select the best tools for the data type. Then we organise and bring this data together in a way that makes it fast, easy and insightful for customers.
That’s amazing and obviously the GPT platforms have accelerated what Yabble does but what have been the challenges you have faced along the way?
We started working with generative AI back in 2020 when most people didn’t know what generative AI was. We’d ask a question of the data, get an answer and people would be like ‘whoa, how did that happen?’. And start looking behind the computer to see if someone was there generating the answer or asking if we had someone in the back room typing! We have had to do a lot of education over the last three years. But today, we now have people coming to us and saying ‘we’ve heard about Yabble, and I really want to use it. How do I bring generative AI into my insight program?’ And it’s happened fast. I haven’t seen the market change so quickly before.
Where do you see Yabble going? What’s next?
The future is really exciting. There are so many different avenues to explore. We are focusing on augmented data generation currently. This is the combining of large language models with other data like review data social media, statistics, demographics and more to create and interrogate virtual personas. We are also exploring multimodal AI to layer voice emotion and computer vision, into our qualitative data products . This allows the AI to get closer to the replication of human senses and allows us to unlock more understanding. And finally, looking at how we can us generative AI to analyse quantitative data using more advanced analytics – like having a data scientist in your pocket. Ultimately, if we can create tools that give all the amazing researchers out there more time to think, then I think we’re all winning as an industry.
Yabble is AI for insights. Yabble creates instant insights so you can take instant action. From the world’s first ChatGPT plugin to a suite of tools that count, summarize and chat with your data. Secure, automated AI solutions for every stage of research – Yabble is the leading AI solution for insights. Supercharge your insights. Yabble it.
At Day One Strategy, we start every day as if it’s our first, full of enthusiasm and a desire to try new things. If you feel the same and would love to leverage the combined power of tech and human intelligence for smarter, faster insight then get in touch: email@example.com