The Gap Between Promise and Practice
For years, enterprise leaders have recognized a fundamental tension in software development: the need to provide engineering teams with realistic data for testing and innovation, while simultaneously protecting sensitive information from privacy risks. Traditional approaches often forced organizations to choose between access and security. Data was either restricted, creating bottlenecks for development and testing, or distributed broadly, increasing compliance concerns.
Synthetic data emerged as a compelling solution to this dilemma. By generating new datasets that reflect the characteristics of production data without carrying its privacy risks, organizations could theoretically increase test data availability while maintaining governance. Recent research indicates that more than half of enterprises (51%) already use synthetic data to protect sensitive information, alongside other methods like static and dynamic masking.
Refer & earn: 30% of a client’s first month, then 10% of every month after that, for up to 24 months per client. See the partner programme →
However, despite this widespread recognition of value, actual adoption remains surprisingly limited. The gap between belief and practice is widening. While enterprise leaders overwhelmingly believe in the promise of synthetic data, adoption lags significantly. According to recent Perforce Delphix research, the biggest barrier to adoption is not awareness, compliance, or governance. It is data quality.
The Quality Barrier
The core issue lies in trust. Across industries, enterprise leaders increasingly recognize synthetic data’s value in protecting sensitive information, moving at “agentic speed,” and expanding data access without increasing risk. Yet, skepticism persists regarding the technical reliability of these generated datasets.
The statistics reveal a stark reality: only 38% of leaders believe synthetic data delivers sufficient realism, and just 34% believe it preserves referential integrity. Referential integrity—the logical consistency between related pieces of data—is critical for accurate software testing. If a customer record references an order, and that order references a payment, those relationships must hold true in the synthetic dataset just as they do in production.
When synthetic data lacks realism or fails to preserve these relationships, development teams cannot rely on it. Many organizations have experimented with synthetic data but did not see improvements in application development and testing because the data lacked context. As a result, enterprises are missing out on the opportunities that true, high-quality synthetic data can provide.
This quality deficit has direct consequences for software delivery. If data lacks realism or does not reflect actual system behavior, teams risk validating applications against scenarios that do not exist in production. The result can be defects, failed releases, and inaccurate testing outcomes. For synthetic data to support enterprise development at scale, it must behave like the systems it represents.
Developers Prioritize Utility Over Protection
A significant disconnect exists between how executives and developers evaluate synthetic data. While executives often evaluate synthetic data through a governance lens—focusing on privacy and compliance—development teams evaluate it through a quality lens. Currently, 66% of software leaders report they are not using synthetic data in development and testing.
For engineers, the primary concern is not whether the data is protected, but whether it is useful. Developers and testers need synthetic data to be more than fake; it needs to preserve the context, relationships, and realism that make data meaningful. A random number in a credit rating field might be technically correct, but it would not reflect real-world data bounds and ranges, rendering it useless for testing complex financial logic.
“Developers and testers don’t primarily ask whether data is protected. They are asking whether it is useful,” the source notes. This perspective highlights why many synthetic data initiatives fail to gain traction within engineering teams. When teams test applications, business processes, workflows, integrations, and customer experiences, they require data that mirrors the complexity of production environments.
The Complexity of Modern Enterprise Systems
The challenge is compounded by the architecture of modern enterprise applications. These systems rarely operate within a single database. Customer information may live in one system, orders in another, and payment information in a third. Testing only works when those relationships remain intact across environments.
When synthetic data tools fail to maintain these cross-environment relationships, teams end up testing disconnected components in isolation rather than real business processes. This fragmentation undermines the validity of the testing phase. Organizations evaluating synthetic data tools must therefore treat specific capabilities as non-negotiable, particularly the ability to customize generation rules to uphold referential integrity across different data environments.
By ensuring that synthetic data maintains realism and utility, organizations can bridge the gap between governance requirements and development needs. This approach allows teams to generate purpose-built data for new applications, features, edge cases, and AI-driven initiatives where production data may not exist or may be hard to use.
AI Acceleration Demands Trusted Data
The rise of AI-assisted development is exposing a long-standing challenge that is now impossible to ignore: software delivery can only move as fast as the data supporting it. While code generation has accelerated dramatically, access to high-quality test data has not kept pace. Synthetic data offers the potential to bridge this gap by allowing teams to create realistic testing scenarios on demand.
This capability is especially valuable as AI agents increase the demand for readily available, high-quality data across the business. However, adopting synthetic data comes down to whether organizations trust that certain solutions are evolving and can maintain realism when they haven’t in the past. A modern, AI-first approach that prioritizes data quality at scale can address these legacy concerns.
Furthermore, synthetic data is not necessarily a replacement for every testing approach. Increasingly, organizations are finding value in combining both approaches: using masked data for known workflows while using synthetic data to expand coverage and test entirely new scenarios. Together, they provide both realism and flexibility, allowing teams to shift validation earlier in the development lifecycle.
What Business Leaders Should Do Next
For business leaders deciding how to proceed with AI and synthetic data strategies, the message is clear. The next phase of synthetic data adoption will depend on whether engineering teams trust the data enough to build, test, and innovate with it. Companies that bridge the trust gap will not be the ones that simply generate more synthetic data. They will be the ones that generate realistic, referentially accurate, fit-for-purpose data that developers can confidently use to speed up their releases and keep pace with AI.
To move forward, organizations should focus on:
- Evaluating Data Quality First: Move beyond governance metrics and assess whether synthetic data tools can deliver the realism and referential integrity required by engineering teams.
- Ensuring Cross-System Integrity: Verify that synthetic data solutions can maintain relationships across disparate databases and environments, reflecting how enterprise applications actually function.
- Aligning with Developer Needs: Engage development and testing teams early to understand their definition of “useful” data, ensuring that synthetic datasets support real-world business processes and edge cases.
- Combining Approaches: Utilize a hybrid strategy that leverages masked data for known workflows and synthetic data for new scenarios, maximizing both security and testing coverage.
Ultimately, synthetic data adoption isn’t an awareness problem. It’s a trust problem, and trust comes from data quality. As software delivery accelerates toward agentic speed, trusted data becomes a competitive advantage. Organizations need data that can be generated quickly, but it must also accurately reflect real-world relationships, business processes, and system behavior.
Faida Workforce supports organizations in navigating these workforce transitions by providing insights into how technology changes impact skill requirements and hiring strategies. Our editorial team focuses on evidence-based analysis to help businesses make informed decisions about their talent and technology investments.
Sources
Only 34% trust synthetic data’s referential integrity — and it’s holding adoption back
30% of their first month, then 10% every month after that
Refer a business to Faida Workforce and you earn 30% of the client’s first paid month, then 10% of every month for up to 24 months per client — rising to 15% from the accrual after you pass 10 active clients. Paid by PayPal, batched on the 1st.
10 Pro clients = $2,598.70 over 24 months — $3,748.13 at the 15% tier.
Figures are potential earnings worked from the published rates, assuming every referred client pays for the whole cap — not a promise, not an average. The $1 trial is never commissionable.
