Blog The Rise of Syndicated Synthetic Data in Market Research Synthetic data has moved quickly from an experimental idea to a much broader part of the research conversation. Early skepticism was warranted: researchers have long had to balance speed and new technology with the human evidence that makes research valuable. What has changed is the range of ways synthetic data is now being created. Some approaches build on structured syndicated research and ongoing category conversations; others draw from individual studies, collections of research conducted over time, or combinations of survey and passive behavioral data. That expansion creates both opportunity and risk. As more organizations experiment with synthetic data, the quality of the source evidence and the methods used to generate synthetic outputs from it become increasingly important. Structured syndicated research is one natural fit for this technology because providers may already have ongoing category-level evidence from which to build. Morning Consult, for example, conducts monthly interviews in multiple countries, then trains AI on those conversations. Subscribers can query the AI during meetings instead of commissioning new research every time a question arises. This approach is gaining momentum. The Qualtrics–Pure Spectrum partnership is exploring similar territory, combining user experience data with profile information to create synthetic research applications. The economics can also be attractive when multiple companies share the cost of building category-level intelligence. That model can give subscribers faster access to existing evidence without commissioning a separate study for every question. Syndicated synthetic data is reshaping insights work in specific ways — how we access category intelligence, when we can get answers, and what questions become worth asking. Getting familiar with these new approaches now puts you in a stronger position as they develop. Where Synthetic Data Can Add Value Synthetic data can shorten research timelines when existing evidence is sufficient to address a new question. Rather than fielding a new study every time, teams may be able to query previously collected and modeled evidence first. This empowers teams to reach refined concepts faster while maintaining quality controls where they matter most. More rigorous validation can then be reserved for the decisions or questions that require renewed human evidence. These efficiency gains can also make it practical to explore smaller questions that may not have justified a standalone study. As synthetic approaches expand, validation becomes more important. Different models may draw from very different foundations: a single study, an archive of prior research, continuously collected survey data, passive behavioral signals, or some combination of them. Researchers need to understand both the quality of that source material and the methods used to generate synthetic outputs from it. That creates a compounding quality risk. Poor-quality sample can weaken the human evidence entering a research system, while unproven projection methods can amplify those weaknesses when synthetic data is generated from that evidence. Evaluating synthetic research therefore requires scrutiny of both the underlying data and the methodology used to extend it. Early-stage testing is one promising application for synthetic data. Faster access to existing evidence can support initial concept screening and rapid iteration before teams invest in more rigorous validation studies. Product concept testing is one example. Researchers can test multiple variations and get real-time feedback during strategic meetings. Advertising creative follows the same pattern. Teams can quickly evaluate different brand positions and directions before committing to expensive production and media buys. So to recap, synthetic data can support: Faster access to existing evidence without new fielding for every question Faster early screening with validation reserved for final decisions Lower costs through shared or subscription-based models More research on smaller questions that may not justify a standalone study Early-stage testing and rapid iteration Faster concept and creative feedback during strategic discussions The Tracking Study Opportunity Syndicated synthetic data is expanding beyond traditional syndicated research boundaries. “I really see a shift coming where there’s going to be less ad hoc survey research and more synthetic data,” explains Frank Kelly, Market Research Practice Lead at Virtual Incentives. One area where synthetic approaches may have particular value is tracking research, where repeated measurement has long created both operational and data-quality challenges. The Problems Plaguing Traditional Tracking Tracking studies are facing serious quality challenges that affect their reliability. Survey exclusions that once helped protect data integrity have become more difficult to maintain. The industry used to enforce 3-12 month exclusions between similar surveys. Those protections have weakened as participation increasingly spans multiple panels and sources. Without these safeguards, the same respondents end up taking too many surveys. For example, they might be asked about Coca-Cola awareness in January, then get the exact same questions in February. Naturally, they’ll recall Coca-Cola more readily the second time, not because of any marketing impact or authentic brand awareness, but simply because the questions are fresh in their memory. This skews the tracking data, which makes tracking studies increasingly unreliable for strategic planning. Beyond the reliability challenge, highly standardized trackers can also struggle to surface insights outside the questions and metrics already built into them. How Syndicated Synthetic Data Offers a Solution Syndicated synthetic data offers a solution. Instead of commissioning expensive individual tracking studies, brands subscribe to category-specific AI intelligence trained on ongoing conversational data. A continuously refreshed source of conversational evidence may reduce some of the repeated-exposure problems associated with traditional tracking studies. Instead of asking participants the same awareness questions wave after wave, researchers can draw from broader conversations about category usage and experience. That advantage still depends on the quality of the underlying sample and the method used to generate the synthetic output. Synthetic data does not remove quality risk; it changes where researchers need to look for it. Conversational data can also create more room for unexpected findings. Because participants are not limited to a fixed set of predetermined metrics, AI-supported analysis may surface themes or opportunities that were not anticipated when the research was designed. “You’re much more likely to find innovative ideas or new ideas that the researcher who wrote the survey might not even be aware of or thought of,” Frank explains. The resulting intelligence may also become available more quickly than a traditional tracking cycle, giving teams another source of evidence when decisions arise. In summary, syndicated synthetic data can support: Reduced repeated-exposure effects through continuously refreshed conversational data Less reliance on repetitive awareness questioning More room for unexpected themes and innovation opportunities Faster access to category-level intelligence Decision-making between traditional research waves An alternative to repeatedly fielding the same standardized questions The Future of Market Research Synthetic data is likely to develop along several paths rather than one dominant model. Structured syndicated research can make category-level synthetic intelligence available across organizations. Other researchers may favor systems built from continuous survey data, passive behavioral evidence, and repeated contributions from the same participants over time. What these approaches share is a growing need for transparency and validation. Researchers need to understand where the source data came from, how representative and reliable it is, how synthetic outputs are generated, and where renewed human evidence is still required. The opportunity is significant, but so is the responsibility. Synthetic data can help researchers make better use of evidence they already have and explore questions more efficiently. Its value will ultimately depend on whether the industry applies the same methodological discipline to synthetic evidence that it expects from human research. What is synthetic data and how is it best used? Synthetic data uses existing research evidence to generate modeled responses, patterns, or insights. Depending on the approach, that evidence may come from individual studies, ongoing conversations, accumulated research archives, passive behavioral data, or combinations of those sources. Its usefulness depends on the quality of the underlying evidence and the rigor of the methods used to generate the synthetic output. What are the main use cases for synthetic data? Synthetic data can support early-stage concept testing, creative exploration, rapid iteration, category intelligence, and some forms of tracking research. The appropriate use depends on the quality of the underlying evidence, the method used to generate the synthetic output, and whether the question requires renewed human input. Can synthetic syndicated data replace tracking studies? Synthetic data may complement or replace parts of a traditional tracking program, particularly when it is built from continuously refreshed, well-validated evidence. It can reduce reliance on repetitive questioning, but its quality still depends on the underlying sample, source data, and projection methods. Share