The FinTech industry used to be very secretive when it came to the transparency of data from analysts. But with the emergence of machine learning and AI, alongside cutting-edge solutions to bridge this gap, analysts and contributors are now held accountable, and layman investors are offered simplified information for more sound and efficient financial decision-making.
On today’s episode of Startups On Demand, I am joined by
Today, we talk about the idea behind the inception of TipRanks, what this software exposed in the industry, the role of ML/AI in the process of simplifying data, and what makes it stand out from other industry players.
Omri: Can you tell us where the idea behind TipRanks came from?
Uri: This was an evolution of an idea. When we started, we had one single feature, which was to measure the performance of cell site analysts. The world of investment was lacking transparency. Retail investors such as myself and my friends were getting their advice online, and as retail investors, we’re very responsive to the content. The idea came from a personal bad investment that I made after following bad advice, which led me to think that there should be more transparency. Basically, we were the first company in the world to measure the performance of what any analyst is saying, then we started using machine learning to measure what every financial blogger is saying. And then we basically started to expand our services until we became the resource platform that we are today.
It is important to say that our goal is to level the playing field for retail investors. Bloomberg is very institutional-oriented, and we have a lot of institutional clients as well. But our goal is that every average person can access the same level of research.
Omri; In terms of machine learning and AI, how do you incorporate that into the software?
Uri: At the end of the day, we’re a natural language processing company. What we do is we analyze financial service websites, financial blogs, and social media, and then we started adding analyst’s research, corporate findings, and whatnot. And a lot of these datasets are very structured, and when it’s structured, we use what is called a “rule-based technology” because, for instance, if I want to understand what an analyst is saying, I know he will always use the terminologies like “overweight,” “underperform,” “reiterated,” “upgraded,” “downgraded,” etc. or a dictionary of about 500 words. When we want to analyze more complicated datasets such as social media or financial blogs, many times it will have grammar mistakes. They can be sarcastic, they can make general mistakes, and they can change their minds – it’s unstructured. In order to analyze it properly, you need to use machine learning which is a statistical algorithm. In order to do that, we need to train our software for years so it can get to the maximum accuracy possible in terms of NLP.
Now, we’re looking into the new revolution of OpenAI to see what we can do with that as well and create real-time content that simplifies complicated stories.
https://youtu.be/4SAmK6mwgYU?embedable=true
Omri: What makes TipRanks different from other players in the market?
Uri: We are obsessed with simplifying things for layman investors. There’s one thing we really disrupted, which is transparency in Wall Street and holding analysts accountable, which was something that no one has ever done before, and we also made everyone hold their contributors accountable.
Omri: What are your thoughts about the industry, analysts, hedge fund managers, etc. after TipRanks was put into action?
Uri: Before we started TipRanks, we thought there needs to be a solution on how good anyone is at giving advice. After starting TipRanks, we realized that the real challenge is making this advice publicly available. It’s so difficult to get one database showing you all this information. No one wants to license us the detailed list of what every analyst is saying, only the high-level consensus, which led us to understand that this is an even bigger challenge than holding them accountable.
In terms of performance, my initial thought was some analysts are good, and some analysts are bad. Some of the best analysts have been outperforming year after year, and if you look at the top 100, there’s a nice list of analysts on the list year after year. In terms of hedge funds, the truth is we can only access a part of the story because they don’t need to report their positions on a daily basis. Also, they are not required to report any short positions that they have.
Omri: Who are some of your favorite big-time investors?
Uri: I’m a big fan of Carl Aiken. I saw a documentary about him, and he’s such a smart guy. I love how active he is in his positions. He really sees things and says them as they are.
Also published here.