Feb
Can AI & Human Researchers Coexist In Market Research?
jerry9789 0 comments artificial intelligence, Brand Surveys and Testing, Burning Questions
AI In Market Research Today
With 90% of the world’s data created in just two years time between 2021 and 2023 and the global data volume standing at 149 zettabytes by 2024, it’s understandable why AI would be readily adopted by the market research industry. Traditional methods of data collection and analysis would hold a place in market research but they simply aren’t as powerful as AI when it comes to handling all that staggering volume of data. But is AI powerful enough to take the place of human researchers?
AI enables research teams to move, process and analyze massive datasets with speed and accuracy, efficiently handling all the repetition and scale involved in the research process. From drafting questionnaires to monitoring survey data quality, from analyzing open-ends to formulating dashboards and charts, AI fully automates the research process leading to faster and better decisions at a scale beyond the capabilities of human researchers.
But is AI the endgame for market research? Does it make human researchers obsolete?
Image: geralt
Cascade Strategies and AI
Cascade Strategies conducted a member perceptions study for a company looking to develop and implement a brand typology. The overall goal of the study was to help them better understand their different customer type’s overall motivations and aspirations for more effective engagement. As part of the study, we conducted an online survey with over 1,500 of their randomly selected members. We then utilized an AI-assisted Self-organizing Map (SOM) to run all the cases recursively, sometimes millions of times, until it optimizes the separations among the groups. The SOM produced a 6-group solution, with each group having a dominant passion that is served well or poorly by the company, ranging from proclivity for deals and new brands to yearning for customization and connection with other users.
The AI has done the heavy lifting of scanning all that dataset, surfacing themes, and summarizing the respondents. It has done enough to structure the story of each group but not enough tell or paint the whole picture.
This is where the human researchers at Cascade Strategies step in. We came up with names for each group that best described their dominant passion, names resonant enough that they not only convey an immediate idea of what they’re most passionate about but makes them fundamentally relatable even if one doesn’t necessarily share the same propensities: Shopper, Seeker, Learner, Sharer, Individualizer and Intellectual.
In isolation, each group achieves the study’s goal of guiding the company on the most effective way to engage with them. Their sum, however, grants the company an overview on how to improve and further develop its platform by considering and introducing new features that matter to one particular group, but would essentially benefit its membership base as a whole when implemented. For example, the Sharer would appreciate increased opportunities to connect and interact with other experts and enthusiasts of the same interests in the platform by making it easier to make reviews and share content.
AI surfaced all those patterns and signals from all that survey data, but it lacked the judgment and context to elevate it into a meaningful and coherent narrative. Human researchers, on the other hand, saw what story can be told from all those themes and by layering in human understanding, they’re able to tie them down to actionable business decisions.
Image: Christina Morillo
Leveraging AI In Market Research
So would AI replace human researchers? We’d like to frame our response to this question with the words of Joseph Weizenbaum, one of AI’s early researchers: “We can count, but we are rapidly forgetting how to say what is worth counting and why.”
Yes, AI is powerful enough to handle large amounts of data to identify patterns, cluster themes, and summarize respondents, but it generates outputs rather than insights. Outputs foster decisions rooted in logic and reasoning, but insights spring from judgment and context. Outputs can provide directions and surface themes from which stories can be framed, but insights take it one step further by asking what matters and why it matters, adding depth and resonance to the story.
In addition, Weizenbaum posits that computer programming can make decisions but it can’t ultimately choose. Just like insights, choosing requires judgment which takes in emotions, values and experience.
We at Cascade Strategies are among a growing number of proponents who believe that AI works best as a tool and extension of human intelligence and talents. AI strips the friction from manual, repetitive work without compromising methodological rigor and accuracy, but rather than adopting it for the sake of automation, we choose to see it as a freeing and empowering agent that enables researchers to focus more on interpreting data with the context of human understanding and values, translating insights into sensible and confident business decisions. Just as quantitative and qualitative research can coexist in the same study, we choose to live in a world where AI and human researchers work together towards the same goal of finding and crafting meaningful and relevant stories worth telling.
Image: Pavel Danilyuk
Featured Image: Ron Lach
Top Image: kc0uvb
Feb
What’s Happening Nowadays With Survey Samples? (Part 2)
jerry9789 0 comments artificial intelligence, Brand Surveys and Testing, Brandview World, Burning Questions
Why The MR Industry Should Start Collectively Caring About Data Quality
In his recent LinkedIn post, JD Deitch offered two explanations on why the sample market is what it is right now and has been for the past two decades: clients either don’t know or don’t care how bad the quality of data sample they’ve been receiving. Between not knowing and not caring, the latter is the more egregious of the two.
In Part 1 of this series, we touched on the challenges presently faced by the sample market: participant engagement, polling representivity and fraud as illustrated by the Op4G / Slice MR scandal. Of the three, fraud captures the most attention, the one that makes headlines, the one that stirs up the most discussion and calls for resolution. The threat of fraud is in everyone’s mind, and that’s why there are measures and protections in place and constantly being developed to detect and address it.
Fraud, however, may not be the key issue out of the three. In the Greenbook podcast, Deitch had pointed to participant engagement as a long-standing challenge that the industry has always been aware of and has tried multiple times to solve. Fraud has always been tagged with large sum of dollars lost or deceptively gained; what most don’t see or take into account is how much revenue or opportunity is missed because of bad or low quality data generated by poor engagement. Yes, fraud undermines credibility and trust in the industry but there always has been avenues to regain them; market failures due to poor data quality may not be as visible but the damage they create linger and influence. And that damage through the decades has now translated to the indifference clients feel towards sample data being produced. As Deitch puts it, “Most survey research projects just don’t matter enough for clients to demand better.”
The current product coming out of the sample market has been commoditized enough that they hardly affect business decisions. Clients don’t see enough value or endorse the same level of confidence in the present product to justify spending more to learn or find out what people are thinking. And if an alternative like AI comes along, clients are roused enough to spend and explore what the other options offer.
“Companies will always want to know what people think. That need isn’t going away.” And this is why renewed focus on the participant experience becomes key. Rather than settle for respondents who have time to fill out questions, find and attract people that are the most invested and involved in the subject matter. Incentivize them for their time and underscore why their thoughts and feelings matter. Connect and foster a healthy yet professional relationship with them. Encourage them to find or refer similar personalities. Build and maintain an engaged panel of quality participants.
New and emerging technologies excite clients and investors alike so look into leveraging them into your methods and processes. Learn the best way to implement AI. Don’t simply deploy new tech to cut costs and time; discover where AI would complement human talent the most and where human supervision is most critical. Collaborate with tech people and developers to design and build systems and applications aligning with your goals and values.
The level of care and effort market research agencies put into their research work would always reflect in the end-product. At Cascade Strategies, we believe excellent and high quality data resonates, and we’re confident it will strike chords in clients to make them care enough. And when clients care enough, they’ll be willing to find out and demand more.
To learn more about how we leverage inspired human thinking with modern cutting edge technology to achieve high quality market research for our clients, please contact us here.
Image: Tumisu
Featured Image: F1Digitals
Top Image: Mohamed_hassan
Aug
Can Synthetic Respondents Take Over Surveys?
jerry9789 0 comments artificial intelligence, Burning Questions
What Are Synthetic Respondents?
AI has increased operational efficiency by streamlining knowledge bases and shortcutting processes so it’s no surprise people and companies are looking for more ways for its application. For market research, one curious consideration is whether it could take over surveys, essentially by replacing actual respondents with synthetic respondents.
Also known as virtual respondents, digital personas, and Virtual Audiences, synthetic respondents are individual profiles constructed by Large Language Models (LLMs) from real or simulated data. Ideally, the data or descriptions used to generate these profiles come from previously conducted surveys and are combined with individual-level demographics, attitudes and behaviors.
Using these synthetic respondents over real respondents could benefit your research with speed, accuracy and cost savings, at least according to their advocates. Basically, you just need to conduct one survey and from the profile description or data you gathered from the actual respondents, you’re able to generate results from the constructed individuals over and over for succeeding studies and research.
Testing Synthetic Respondents
While synthetic respondents could accurately represent real respondents, relying exclusively on the results from these AI-based individuals may not be entirely beneficial. A webinar hosted by Radius Global took a closer look at the potential of AI-generated synthetic respondents through three case studies of quantitative concept testing, quantitative communications research, and qualitative communications research.
Aggregate results for the concept tests involving game controllers indicate somewhat strong similarities between the results of the real and synthetic respondents. This extends to the results from the quantitative communications research when it comes to the believability of statements on the benefits of milk, although there were some differences. The differences were much more pronounced though when it comes to surprise over the same statements, and there was incongruence when considering how each statement could possibly increase milk consumption.
The qualitative communications research was seeking in-depth insights into women’s needs, perceptions, and preferences for running a race or marathon, with the feedback gathered meant to be used for creative content. Personas were constructed from the profiles of six women aged between 18 and 64 years old who ran at least once in an average week. They had an LLM assume each persona to allow a comparison between findings from real participants to synthetic respondents.
They found that while both real and synthetic respondents have somewhat similar responses when it comes to functional aspects as goals for women in general pursuing fitness, the AI responses lacked emotional expressions. There are also little differences in the synthetic respondents’ responses despite having different profiles, and there was even a lack of subtle differences.
As for concerns among women who are aspirational marathon runners, the synthetic personas were consistent in their responses while the real respondents provided more nuances, variety, and perspectives more prevalent among women.
Synthetic Respondents vs. Real Respondents
Synthetic respondents appear to be useful if you’re evaluating existing ideas and concepts; however, if you’re looking for “breakthroughs” or essentially new insights you would’ve never arrived at had you not performed the case study or research, you would need to engage with real respondents, relying exclusively on their results or combining them with that of synthetic respondents. Yes, there could be cases where synthetic respondents could be used, but the results must be extensively validated. It would also require increasing the efficiency of how data used to construct these individuals are analyzed in addition to enhancing the quality of the data and information gathered for these profiles through thorough screening, intelligent probing, and smart choice models.
There is a place for synthetic respondents in market research, but as another tool in a researcher’s toolbox. They won’t be taking over surveys or replacing actual respondents wholesale anytime soon, it seems, as that elusive “Eureka” moment researchers seek is inherently tied to the nuances and perspectives of human emotion and experience you simply can’t construct.
Photo courtesy of Pavel Danilyuk
Nov
“Humanizing” Market Research with AI
jerry9789 0 comments artificial intelligence, Brand Surveys and Testing, Burning Questions
The Boon and Bane of AI
The increasing and widespread utilization and demand for Artificial Intelligence have been met with both excitement and reservation. Excitement for the possibilities AI’s implementation unlocks, oftentimes steps ahead of the curve or beyond expectations; reservations not only stemming from the risks over its unethical and unchecked use, but also the ramifications for human involvement now that intelligent machines represent an optimized and economical choice for completing tasks and processes. But can there be a middle ground somewhere where AI and human engagement coexist and collaborate?
The “Humanization” of Market Research
“Capturing the Human Element in an Artificial World” by Eric Tayce (Quirk’s Marketing Research Review, Sep-Oct 2024) posits that such a midground is possible, especially in market research. An industry that’s all aware of its excessive dependence on technology to necessitate a push to “humanize” research data, it saw a dramatic shift from “data-intense tomes and clinical-sounding slide titles” to “streamlined, narrative-style reporting” focused on “the unique motivations and experiences that drive customer behaviors.” The latter “humanized” approach is able to communicate business goals while connecting and engaging on an emotional level. However, generative large language models (LLMs) cast a shadow on this “humanized” approach by offering synthetic outputs and progressive algorithms.
But combining both AI and efforts to “humanize” research can result in the whole being greater than the sum of its parts. The article shared that AI can help collect more unstructured data from survey research by employing conversational chatbots to create a natural, richer experience for the respondent. That unstructured data in turn can potentially provide more organic, more human insights with AI-powered algorithms, an undertaking that was once considered too complex or time-consuming. AI can also build multifaceted perspectives through context by linking survey records with a broad range of data sources. And in lieu of traditional static deliverables, data and insights can be presented in a vibrant and interactive narrative by an AI-powered persona.
The “Human” Element
All these interesting prospects can only be achieved when AI is tempered by high-quality human input and thoughtful implementation considerate of ethical and moral implications. Aside from AI mistakes and hallucinations existing, AI has been observed to be too helpful and excitable. Human oversight and input remain key in ensuring AI models are trained, fine-tuned and grounded with quality and relevant datasets while having enough flexibility to engage appropriately in open-ended interactions.
There’s no denying just how transformative AI is in reshaping industries today, including market research. Despite concerns of machines taking over jobs, one can look at it with the perspective of roles changing and adapting. AI with its generative and synthetic capabilities can elevate the “humanization” of market research, but to get to that point we simply can’t forget that humans are indispensable to the whole process.
Featured Image Copyright: GrumpyBeere
Top Image Copyright: Darlene Anderson
info@cascadestrategies.com
2032 Newport Way NW,
Issaquah, WA 98027
(425) 677-7430