Categories
Introduction
Across the market research industry, more and more organizations and companies are using AI to speed up audience insights analysis. The reason is simple: AI can streamline previously lengthy processes. Formerly time-consuming work can now be done in seconds, with human beings on hand to ensure accuracy.
AI tools save time and money but they also offer restech organizations another possibility: synthetic data.
Synthetic data is a means of simulating respondent behavior, attempting to create a model based on previously-divined insights. “Effectively,” says Lindsay Fordham, SVP, Product at Cint, “you’re asking LLMs to respond to data as if it were supplied by a human being.”
As the name suggests, synthetic data is data that has been algorithmically generated from previously captured real world data samples.
“There’s a lot of opportunity in this area to explore how LLMs can help us in research, and filling gaps in our data sets to help build a bigger picture,” Fordham says. “But we’re never going to replace the need to collect data directly from consumers. We absolutely still need strong training data sets and human opinions to feed the machine. Otherwise we will be subject to analysis that’s based on assumptions of what human beings think versus what human beings actually think.”
Wanting to understand more about market researchers’ confidence in AI and the use of synthetic data in market research, Cint devised and deployed a survey on the topic.
Survey methodology
In mid-October 2024, almost 200 market researchers and insights professionals from Cint’s customerbase were invited to participate in a research project. They were asked to respond to a survey regarding their usage of AI and synthetic data, and their outlook on those shifts in the Restech landscape.
All respondents were required to be active customers of Cint’s software, and respond to the survey within a week.
The respondents spanned the world and industries from market research to education and brands, and job roles ranged from CEOs to account managers.
AI usage and synthetic data
Cint asked respondents to report on where AI comes into play, and how much they rely on it.
Our survey results showed that AI plays an ever-larger role in most of our working lives. Products and services like ChatGPT and Google Gemini have, for many people, become widely-used tools in our 9-5. Nearly three quarters of respondents reported (74%) using AI-powered software in their current job roles.
When it comes to synthetic data, respondents were then questioned about their use of synthetic data in the following categories:
- Increasing sample sizes for niche targeting (“boosting”)
- Asking additional questions to an already completed survey, filling the answers with synthetic responses
- Reducing Length of Interview (LOI) by adding synthetic answers to unresponded questions
- Turning incompletes into completes by adding synthetic responses to unanswered questions
- Replacing human respondents entirely in certain types of projects
Respondents also reported strong feelings on synthetic data, reinforcing the need for transparency and greater communication.
Fordham says that close communication with customers is crucial for ensuring that everyone’s on the same page when it comes to predicting what will stick and what’s another soon-to-be abandoned buzzword in the world of market research and the insights industry.
“We are talking to all of our largest customers about how we could collaborate, and we are very open to doing proof of concepts.”
To read more about the role AI and synthetic data plays in our respondent’s professional lives, head here to download the report in full.
