Sramana Mitra: I did a story some time ago on a company called DiscoverOrg. They were mapping out the org charts inside of organizations to sell to sales organizations. They have since been acquired by ZoomInfo. As you are talking, I am thinking that that kind of data layered onto what you are doing would be very valuable. Is that on your radar?
Oleg Rogynskyy: We acquired a company called ClosePlan in October 2020. They enable companies to build org charts with a prospect inside their Salesforce. This is a much newer version of DiscoverOrg. We have seen DiscoverOrg disappear off the radar since they went public because before going public they were scraping LinkedIn to power their data set.
That is a gray area today based on all the lawsuits. They cannot do that anymore. In our case, we make org charts collaborative. About 100,000 salespeople are collaborating on org charts within their Salesforce. We are the company in between that makes everybody’s org charts automatically without having to scrape on LinkedIn or anything like that. We found a better way to do that.
Sramana Mitra: Is that proprietary? If one organization that has a large number of salespeople are building out the org chart for themselves, do you take that knowledge and let other people use that or is it limited to the salespeople from that organization?
Oleg Rogynskyy: They have control over it. They can keep it to themselves, but then they will not get other org charts built by someone else. They can share theirs and get everybody else’s.
Sramana Mitra: You are learning that people are more interested in collaborating than being proprietary.
Oleg Rogynskyy: We have 150 large enterprises on the platform who are all collaborating.
Sramana Mitra: Okay, got it. Super interesting. I love what you are doing. I think that it is a great time to do something like this because of all the abundance of data. It sounds like you did a very good job of cold starting the data leverage process. Congratulations.
I have one last question before we close. Form your vantage point and with your skillset, knowledge, and experience, what problems or what space would you go after if you were starting a new company in the AI domain today? The point of view in which I am asking you this question is that other entrepreneurs who are reading or listening to this interview are wondering about what problems they should solve.
Oleg Rogynskyy: It all comes back to activity data. Let me explain why I think that. To build an AI model of doing something better, you need to have a pretty high-resolution description of how it has been done before and whether it has worked or not. Collecting activity data of players of a specific market, be it salespeople or wealth managers is collecting all telemetry. To train machine learning models is the path.
I don’t believe that it is possible to build a large company just based on better machine learning algorithms. You can only build a large company these days by having a proprietary data set that nobody has access to. If I were to advise entrepreneurs on where to start companies, I tell them to identify an area of human activity where there is no massive public company optimizing that activity for a better outcome.
Find a way to collect the proprietary datasets of humans in that domain. It can be how people buy clothes, how people buy airplane tickets and others like that. You can then build a company that collects activity and optimizes behavior for better outcomes in the domain. There is a lot of white space there.
Sramana Mitra: Great. I thoroughly enjoyed this conversation and I hope to keep in touch with you and learn more about how you are progressing. Good luck with everything.