Among the many transformations that took place as a result of the global pandemic was disruption of traditional data sources.

If a restaurant suddenly had to suspend operations for six months and needed a loan, how could a lender rely on profit and loss statements or tax returns to evaluate the risk?

This is the kind of puzzle that plays into the mission of Explorium Inc. The company, which was founded in 2017, has developed a platform that leverages search capability from a large amount of external data sources and discovers the most relevant information to feed analytics and machine learning models.

In the case of a closed restaurant, this might include data around previous foot traffic, online reviews or even delivery services.

“We have seen the whole data market changing in the last year and a half with the pandemic coming in,” said Ajay Khanna (pictured), chief marketing officer of Explorium. “The models that the industry was working on for evaluating loan applications were not working anymore. Many of our customers had to depend on external data to make those decisions.”

In anticipation of the AWS Startup Showcase: The Next Big Thing in AI, Security & Life Sciences — taking place on June 16 — Natalie Erlich, host of theCUBE, SiliconANGLE Media’s livestreaming studio, spoke with Khanna for a special CUBE Conversation. They discussed the challenge of searching and mining for relevant data across the Web, how Explorium has designed a platform to extract necessary information, forthcoming changes by major online content and technology platforms that will restrict publicly available data, and the company’s recent fundraising success.

Finding insight in data mass

The problem with reliance on external sources of data is that it can add up to a mind-numbing amount of information. The preponderance of internet information has become so vast, no human can possibly sift through it all, much less analyze it for relevant content that pertains to a specific business.

“The big challenge is that getting external data is really hard,” Khanna explained. “It is hard to access, and you don’t even know how many data sources are out there. If you go to Data.gov, there are over 250,000 data sources there.”

To help with this challenge, Explorium uses machine learning to automatically extract, aggregate and integrate relevant data characteristics from external sources and match those with a customer’s internal datasets. The resulting analytical and machine learning models can understand buyer actions, forecast demand or assess risk.

“There’s tons of data out there on website interaction, which is not within your organization, but you want that data to get a better understanding of your customers,” Khanna said. “We provide an external data platform to discover thousands of relevant data signals that you can use in your analytics or machine learning models. You can have access to thousands and thousands of data signals, and you can take those and enrich your internal data, do transformations, and then build pipelines that business analysts can use.”

New restrictions coming

Businesses that may have previously relied on the use of third-party cookies, or small pieces of code dropped into a user’s browser engine when a particular website is accessed, may soon find a narrower path to insight. In January of last year, Google LLC announced it would no longer allow third-party cookies on Chrome, the most widely used browser in the world.

In addition, Apple Inc. has tweaked its more recent iPhone iOS updates to allow third-party tracking across various website apps only if the user grants permission first. This will likely require firms to seek new solutions such as Explorium’s for the agility needed to compete.

“There are new transparency requirements that you have to abide by,” Khanna noted. “If those signals are gone, how do companies better understand their customers? Having that agility will determine the competitive advantage for these companies.”

Explorium has been gaining its own market traction of late. The startup closed two funding rounds over the past year, a Series B for $31 million last July and a $75 million Series C in May.

“Companies are realizing that just having machine learning algorithms is not enough; everybody has the same algorithms,” Khanna said. “The advantage is the data you have and the domain expertise that you have. The use of external data is going to increase with time, and that increased interest is what we are seeing here.”

Watch the complete video interview below, be sure to check out more of SiliconANGLE’s and theCUBE’s CUBE Conversations, and tune in to theCUBE’s live coverage of AWS Startup Showcase: The Next Big Thing in AI, Security & Life Sciences on June 16.

Photo: SiliconANGLE