Fast paced digital innovation is one of the key themes in Financial Services in 2021. COVID-19 has further driven the importance of digital and the need to innovate, faster and more effectively. However, many of the large financial institutions remain hamstrung when innovating. Their Achilles heel? Data.
We believe synthetic data represents a revolution in testing and offers the potential to turbo change innovation.
Synthesized data assets are totally new data sets that look, feel and behave like the original data set but are not replicating any of the detailed data itself. The data can be shown to have the same characteristics, signals, attributes and ranges without exposing the original data set beyond the synthesis point in an organisation. Once created through the Synthesized engine, the data can be placed into user acceptance, test, development or sandbox environments with no risk of data leakage compromising the original customer or corporate data assets.
The benchmarking process will indicate any concentrations, biases, correlations and gaps in data providing scores and indicators. The Synthesized data assets can then be treated with this as a known, or rebalanced to eliminate or smooth out any of these features.These new data assets address the problem of volume and variety for sparse, nonexistent or difficult to get data. Synthesising the data allows speculative uses to be investigated and validated before committing to the time consuming processes of data owner approvals, cross-border or cross-division approvals and legal approvals. Removing the cost, time and risk from the data sharing enables perspective and opportunistic innovation.
It is now possible to build data sets many magnitudes the size of the original test data. This allows for volume and scale testing of systems or models using representative data which may not have been encountered before but is possible.