The Real Potential of Synthetic Data in Market Research: Your Questions Answered

After our recent webinar on synthetic data, we received some great questions from attendees. Below, we’ve compiled the most frequently asked ones, along with our answers, to help clarify what synthetic data modeling can (and can’t) do in the context of market research. 

Missed the webinar? Watch the full session below:

❓Can synthetic data modeling answer a question that you have not directly asked to the sample?

Short answer: No.
Think of it as very advanced weighting, using different techniques but achieving the same overall result of building on existing data to show it in a new way. Algorithms are trained using actual data – they can extend and simulate relationships within that data, but they can’t generate answers to questions never asked in the first place. Think of it as advanced weighting, not mind-reading. Real respondent input remains essential.

❓Are there clear use cases where NOT to use synthetic data/audiences?

Yes – particularly with 100% qualitative data, extremely small base sizes, or when interactive/subjective content is involved. As a rule of thumb: if weighting isn’t appropriate, synthetic modeling likely isn’t either.

❓Are there any specific demographic, sociographic or other variables that tend to be more critical in the accurate modeling and imputing of data? If so, what are the key variables that need to be incorporated?

Not enough formal testing exists yet. But more inputs – especially demographics and behavioral traits – lead to more robust modeling. Psychographics are promising but still exploratory.

Want to go deeper? The questions above just scratch the surface. We compiled even more of your nuanced, nitty-gritty questions into a free downloadable resource.

👉 [Download the full Q&A document here] and explore the real-world boundaries (and breakthroughs) of synthetic data in market research.

Let’s keep the conversation going. Get in contact with Kyle Hope, khope@questmindshare.com and/or Scott Worthge, sworthge@questmindshare.com