How it works

The Statice software & its implementation

How Statice works

Statice enables your data-driven use cases to be implemented easily and without complications. While putting data privacy first, you can leverage the synthetic data provided by Statice for a wide range of applications.

Synthetic data can be used for statistical analysis, just like original data can. The difference is that synthetic data is legally compliant.

Statice can be easily deployed as an on-premise solution or within your cloud infrastructure. The solution provides you with guaranteed privacy-preserving synthetic data.

How to integrate Statice in 4 steps

The software is installed on your infrastructure

The software trains a generative model on your customer data

Which generates privacy-preserving synthetic data

The generated synthetic data is automatically validated for privacy compliance


We have the answers

What is synthetic data?

Synthetic data generated through Statice consists of entirely artificial and new data points and is guaranteed anonymous while resembling the statistical properties and structure of the original dataset. It possesses the following attributes:

  • Properties and statistical information of the original data are mirrored in the synthetic data
  • The data structure of the original data is retained unaltered in the synthetic data
  • It is impossible to identify real individuals in privacy-preserving synthetic data

What can my company do with synthetic data?

You can use the synthetic data for any statistical analysis that you would like to use the original data for. In essence, the idea of synthetic data is to be used just like real data - with the only exception of not providing sensitive information on real individuals.

Is synthetic data provided by Statice GDPR compliant and anonymous?

Yes, according to Data Protection offices in Germany, synthetic data with a privacy guarantee (such as provided by us) is seen as anonymous data and thus falls outside of the scope of the GDPR.

Can Statice guarantee the anonymity of the synthetic data?

Yes, we can. At the core of our software, we make sure to fulfill a mathematical privacy guarantee, which is called Differential Privacy.

How well does Statice's synthetic data capture the quality of the original data?

This depends heavily on the types and volumes of data being synthesized, and the degree of privacy required, and ranges from extremely well (large data set, tabular or relational data) to not well (small data set, high dimensional). We have shown to capture even hidden features in highly complex and high-dimensional data in the past.

How does Statice capture possibly unknown correlations in the original data?

This is part of our modelling. Our algorithm should capture even "hidden" correlation, i.e. relationships not directly visible from a first exploratory analysis.

How is our solution shipped?

Our solution is usually provided as an on-premise SDK or as a standalone on-premise application.

Do we provide an on-premise solution?

Yes, our software is designed to be used on-premise and thus allows you to process your data locally without it to ever have to leave your infrastructure.

Wondering if this is for you?