Feature
For digital marketing leaders struggling to execute marketing strategies amid data privacy regulations, synthetic data can help with secure and efficient data anonymization.
Synthetic data is generated by applying a sampling technique to real-world data or by creating simulation scenarios where models and processes interact to create completely new data not directly taken from the real world. This data can also be used to help with innovative advertising experiments and for the optimization of product solutions.
Synthetic Data Needs a Central Data Platform
“Synthetic data could have far-reaching implications across the organization,” said Mike Froggatt, senior director analyst in the Gartner marketing practice, who explained there are three main opportunities around synthetic data for brands.
The first is the expansion of privacy-safe customer data models for targeting, the second is data-driven creative capabilities, which include automation for video, audio and display ads, and third is the increased availability of AI-driven models for activities like customer service, product development and training.
He said organizations looking to leverage synthetic data require a central data platform, whether a simple database or specialized technology like a customer data platform (CDP), to house both sample data and the synthetic data created via generative AI tools.
“The database should be able to ingest various points of data from owned, partner and paid customer-facing touchpoints, such as your website, sales or retail partners, and engagement data from paid media, while also easily accessible to export for activation with your media, AI and other partners," Froggatt said.
Among the other tools he recommends are generative AI tools to create the synthetic data and future scenario and modeling software, which can include anything from virtual worlds to digital twins of customers/products.
“Organizations will also want to have measurement benchmarks to gauge the reliability and accuracy of their synthetic data,” he added.
Elise Devaux, head of marketing at Statice, a developer and synthetic data solution specialist, explained synthetic data can be used by brands and marketing departments to augment, rebalance, or create customer datasets that would otherwise be incomplete or restricted due to data privacy regulations.
“This can be especially useful for companies performing customer analytics, simulating campaign scenarios, or testing marketing strategies, which require large amounts of high-quality data,” she said.
For example, brands can train machine learning models on synthetic data to improve the targeting of ads to specific audiences based on factors such as demographics, interests and past behavior.
“Synthetic data also allows valuable first and third-party data sharing among marketing stakeholders without risking customer privacy or compromising enterprise security,” Devaux said.
Related Article: The Role of Synthetic Data in Emerging Digital Workplace Technologies
Synthetic Data Deployment Needs Cross-Functional Teams
Froggatt advised setting up a cross-functional team to brainstorm areas where you have limited data that could potentially benefit from the creation and analysis of synthetic data.
“This team could be comprised of individuals from marketing, IT, legal and sales departments, preferably with experience in product, security and risk management,” he explained.
Devaux said that to properly leverage synthetic data, marketers also should educate themselves about the opportunities but also limitations of this technology.
“To get started, engage your data and analytics team in a conversation around the use of synthetic data, define use cases where synthetic data would bring the most value to your organization, and find the right tool that is capable of supporting your needs,” she said.
Another important best practice is to identify potential use cases, including the use of synthetic data to replace first-party data, which Froggatt noted is notoriously hard to collect and maintain consent.
He notes synthetic creative and media technologies can also lessen dependence on costly productions and be used to supplement the huge amounts of data required to develop accurate AI- and machine-learning algorithms.
Learning Opportunities
Webinar
Jun
6
The Future of Social Media: Emerging Trends and How to Stay Ahead of the Curve
The impact of emerging technologies like AI on social media
Webinar
Jun
7
The Building Blocks of Personalization
This webinar will explore how personalization can create powerful ecommerce shopping experience.
Webinar
Jun
8
Attracting—and Keeping—Your High-Net-Worth Investors
Join PwC to dissect their recent survey findings on a specific customer subset — high-net-worth investors.
Webinar
Jun
13
The Future of Contact Centers with AI at the Helm
Learn how to balance human intuition and interaction with the potential uses of AI to create delightful and lasting CX
Webinar
Jun
20
3 Steps to Unlock Your Customer Data and Drive Revenue
Discover how Twilio Segment’s #1 Customer Data Platform helps implement data-driven strategies.
Webinar
On demand
AI for Marketers: Optimize Personalization Across the Funnel
Understand the benefits of utilizing AI across your entire marketing stack
“Synthetic data should resemble the data that you already have, without any personally identifiable information attached to it,” Froggatt explained. “The beauty of synthetic data is that it should resemble the existing data that your organization has, so it shouldn’t require new skills to analyze or activate it."
From his perspective, the hardest part of synthetic data is acquiring it, typically done by providing a sample data set to a generative AI partner and outline the additional attributes that you’re looking to acquire.
“This is where digital twin of the customer work can come in handy, as well, where you can start to build data from virtual customers in a privacy-safe way," he said. “This can also help test things like customer journey and user experience of your website and other owned properties.”
Devaux said brands can choose from a variety of open-source and commercial synthetic data solutions that provide the machine learning and AI frameworks needed to generate synthetic data.
“These solutions often include tools for evaluating the quality and privacy of the synthetic data, making them an economical and scalable option,” she said. “However, companies can also develop their own approach and toolset by exploring existing deep learning and generative models.”
Related Article: Data, Data Everywhere, But Not a Drop to Drink
Synthetic Data Presents CMOs With New Opportunities
Devaux pointed out synthetic data presents marketers with the opportunity to rethink their approach to leveraging data.
“Over the past decade, marketing trends have emphasized personalized and individual targeting, which relies on collecting and analyzing personal and behavioral data," she noted.
However, customer preferences and data privacy regulations are making this approach increasingly challenging and outdated.
“Synthetic data offers a way for marketing leaders to reconcile innovation, privacy and boost customer satisfaction," Devaux said. “Insights can be uncovered without using sensitive personal information.”
She added that as data privacy laws become increasingly stringent, marketers must find alternatives that ensure customer privacy while facilitating data-driven initiatives.
“Synthetic data represents an opportunity to capture both data-enabled insights and customer privacy,” she said.