San Francisco-based Okera Inc. is on a mission to tackle the barriers that make it difficult for enterprises to take full advantage of their data.

The startup, which was founded by two former Cloudera Inc. executives and is backed by $14.6 million in funding, today exited stealth mode with a platform designed to ease analytics projects that use information from multiple sources. It does so by providing a centralized interface for managing who can access what records and how.

This task is often a challenge for large organizations with a lot of data to govern. In any analytics project, a company must take precautions to ensure that information is used appropriately. The limitations placed on data access can’t be overly strict, since that might prevent employees from carrying out their work. But an organization can’t afford to be lax, either, since insufficient controls have the potential to open the door to information misuse and create legal risks.

“The friction in management and governance is high,” said Amandeep Khurana, Okera co-founder and CEO. “People have to strip out data consumers shouldn’t see, and that means it can take weeks to get access to one data set.”

The problem is particularly pronounced with data lakes, which often contain large amounts of unstructured that isn’t defined by the organizational blueprint called a schema. “The challenge of fine-grained access control and governance is much bigger in a data lake,” Khurana said. “It looks like a mainframe but doesn’t have the sophistication of database management.”

Okera said its centralized approach makes this delicate balancing act easier. The startup’s platform enables administrators to view their companies’ data assets in one place and set restrictions on how they may be used.

Okera says its asoftware can enforce rules all the way down to the individual cells in a database. A bank, for example, could allow its business intelligence team to run analyses on transaction logs without giving them access to certain sensitive details.

“We sit between the data store and whatever tool you’re using,” Khurana said. “We give you connectors for different kinds of data stores, and you can also build your own.” Standard application program interfaces enable developers to plug in scripts written in Python and R, which are popular languages for data analytics and machine learning.

Okera also provides a number of other data protection features designed for more specialized use scenarios. Among them is an anonymization tool that can mask any personally identifiable information in the records being analyzed. Removing employees’ ability to match sensitive data to specific people is not only a handy means of improving privacy, but is an outright requirement in certain cases.

Once the information is ready, companies can make it available to employees via their preferred outside analytics tools. A monitoring dashboard enables administrators to keep an eye on user activity and ensure that everything is done in accordance with internal policies.

Okera claimed its platform has so far been adopted by Capital One Financial Corp. and several other Fortune 500 companies. The startup also boasts an impressive lineup of investors that includes Bessemer Venture Partners.

Paul Gillin contributed to this story.

Image: Unsplash