Discussion groups can be an important marketing tool, offering real information directly from customers’ mouth. However, they also have speed, cost and scope limits. Fortunately, a solution can be on the way: the synthetic tests of the public.
Synthetic audience tests involve the creation of digital twins or customer avatars who can answer questions from marketing specialists.
“Public synthetic tests would allow me to create this audience using real data,” said Camille Manso, partner of the innovation consulting company, Silicon Foundry.
Unlike synthetic data, which is artificially generated, synthetic audiences are built from real data such as demography, purchasing behavior, etc.
“You might ask them for five W,” she said. “Why could you look for a new supplement?” When do you buy it? When would you incorporate this into your daily routine? How could you consider the product?
A “person” contains multitudes
Each “person” in the public represents a different demographic group and not an individual. Thus, they could be sorted by geography, income or everything you have of data.
“To create these profiles, it will still be largely to the definition of man what the segment consists of,” said Manso.
Dig more deeply: Find the ideal point between relevance and discovery in B2B Lead Nurturing
As with everything in marketing, many will depend on the quantity and quality of the data used to build the public.
“You must be prepared today to take advantage of the opportunities that will be available in the coming at three years,” she said. “So what data do you have at your disposal?” To what extent is your data organized? Because these models and tools of AI will not be as good as your data. »»
At the development stage
Currently, audience synthetic tests are “exaggerated” technology, although with enormous potential. It could replace discussion groups and surveys or complete them. These methods are not sufficient by themselves. The discussion groups take time and expensive, and a small sample size can distort the results. Consumers are less interested in participating in the polls and surveys and are not always truthful when they do.
In comparison, public tests of the public could reduce time to information. Instead of weeks spent recruiting, leading and analyzing discussion groups, marketing specialists would be able to ask their synthetic audiences and receive answers immediately.
“You can’t iterate quickly if you are running a discussion group,” said Manso. “While if you talk to an AI and you are behind a computer, you change things when you want. Like: “What if we put it in a white packaging? What if we put it in sustainable packaging? What would be your reaction? »»
Easy to test
Manso suggested that companies can quickly test the usefulness of a synthetic audience by submitting the same questions with it and traditional discussion groups.
Currently, the emphasis is placed on the grouping of individuals based on models of large languages formed on specific user data rather than obtaining a real understanding at the individual level or a “segment of a”.
For the future, a key objective of the synthetic tests of audiences is to nail the microsegmentation, which aligns the broader trend of hyperrsonalized marketing. Interestingly, the synthetic public could be used to understand when personalized marketing becomes too hyper by evaluating what consumers could perceive as “frightening”.
Dig more deeply: Are A / B tests dead?
👑 #MR_HEKA 👑
