AlphaSense cofounder Jack Kokko, WG’08 talks with Karl Ulrich about how his days as an analyst—and as a Wharton Executive MBA student—inspired him to create an intelligent search engine for business.

A Big Problem for Analysts

As an analyst at a big Wall Street bank, Wharton Executive MBA grad Jack Kokko spent a lot of time digging through information, and he vividly remembers the feeling of “fearing that you’re missing something critical that’s putting a billion dollar deal at risk.”

Other analysts need no longer suffer as Jack did, thanks to Jack himself, CEO and founder of AlphaSense.

In Jack’s words, “AlphaSense is an intelligent search engine for knowledge workers in corporations and financial firms. We’re doing to business information what Google did for the internet. We’re organizing it and making it really fast and easy to find the information you need as a business professional.”

A Wharton Solution

The idea for the company came to Jack and his co-founder, Raj Neervannan, also WG’08, when they were doing a class project “very similar to my analyst days,” and saw that, “this problem still was there, that it was still incredibly hard to find information. It was very manual. All of these search technologies and other machine learning, AI, other technologies that had come about for many consumer applications, you had consumer web search engines and other things making things incredibly easy to find information in the web, but none of this really applied in the business world for finding business information.”

How Does AlphaSense Search?

With AlphaSense, you run a search much as you would on Google, but “what comes back is information that’s relevant for a business use case.” AlphaSense searches not just for your exact search term, but for business or financial synonyms as well, so “you don’t have to think about what words a company executive might be using in talking about the concept your searching for.”

Jack explains: “We have hundreds of algorithms that are reading every line of text and categorizing information. So our system knows what company or companies are being talked about in a given document, what industry it’s about based on the language, and many other things that help you quickly narrow down your search to a given industry, to a specific set of companies.”

But Is It A Normalized Data Set?

The results contain “content from thousands of sources, many of them licensed.” That content includes research from many Wall Street firms, including the meta data that analysts add, “but then we also apply our own AI algorithms on top of it. And our algorithms kind of do a second layer of analysis on the same kinds of things because when you have thousands of Wall Street analysts tagging research, some are doing a great job, some are doing a poor job.”

Finally, “to have kind of a really normalized data set where you can rely on that one search engine going across all these thousands of sources, you have to have everything normalized. And so we add the algorithmic analysis and tagging on top of it to correct any human mistakes or misses.”

An Information Edge

Unsurprisingly, AlphaSense has been met with cries of delight from the analyst community—once they learn about the product. Jack and his team still have to go out and sell AlphaSense, because his customers, “they’re not searching for that information edge, because they didn’t know that it was even out there, so you kind of have to go out and proactively reach out to them.”

“Today we have hundreds of hedge funds and other investment firms using the product,” Jack says. Plus they’ve raised around $35 million in VC funding, according to Crunchbase.

All because Jack and Raj thought their Wharton homework should be less work. For today’s students, if they’re using AlphaSense, it is.

Posted: May 27, 2018

Related Content

Read More Stories