One of the most important functions of any modern financial institution is conducting valuations. On Wall Street, this means developing financial models – projections of how a company will perform based on a set of assumptions. Get these models right, and suddenly you can make trades on public equities that bring in enormous profits. Blow it, and see billions of dollars evaporate.
Despite the importance of financial models, the process behind modeling remains as cumbersome as ever. Since today’s analysts develop models using Microsoft Excel and collaborate through email, it is difficult for multiple analysts to work on the same document at once. Even more worrisome, assumptions and models are rarely version-controlled, which means that mistakes have a tendency to amplify given the frenetic pace of a trading floor.
This is the world that Justin Zhen and Gregory Ugwi discovered when they graduated from Princeton together and joined Wall Street in the heyday years before 2008. Now, as founders of Thinknum, a platform for financial analysis currently in beta, they hope to dramatically improve this state of affairs by offering better access to key data and a significantly improved environment for sharing and collaborating on models.
The founders, though, are not limiting their vision to just a few analysts in Manhattan. Instead, they believe that better collaboration around models could do for finance what GitHub did for programmers – democratize finance by allowing anyone, anywhere to contribute to a better understanding of valuation and become recognized for their work.
Remodeling Financial Analysis
Right now, Thinknum has two products: Cashflow Models and Plotter. With Cashflow Models, users can develop their own valuation models using an Excel-like product for the web. Similar to GitHub, once a model is built, analysts can openly share it for free, or securely save it for a price.
“Analysts often spend a lot of time rebuilding models again and again, but what happens if we could fork it?” Zhen asks. At any time, a user can take any model and edit its assumptions, parameters or formulas, allowing them to experiment with different approaches.
Data is fundamental to modeling, and traditionally, getting high-quality financial data has proven difficult for startups without the resources of Bloomberg or Thomson Reuters, the incumbent financial data providers. That changed recently, as the Securities and Exchange Commission has begun requiring financial disclosure forms to be submitted in XBRL format, the eXtensible Business Reporting Language.
Thinknum takes advantage of this new environment by allowing instant access to company financial data. Since the variables are standardized, this means that you can run the same financial model across an entire industry in a matter of moments. Or, using a tool called QuickBuilder, you can use sliders to adjust such variables as gross margins and revenue growth to evaluate the sensitivity of a company’s valuation.
One issue that modelers face is finding the right index data to compare a company to. For instance, modelers may want to compare a company to the rate of expansion of the manufacturing industry (using an index like a Purchasing Managers Index) or growth in the U.S. GDP. Through Plotter, analysts can evaluate correlations between data and use a visual tree to discover indices.
“In current tools that [analysts] use, they need to know exactly what they want, and they don’t see relevant data that they haven’t thought of,” Zhen says.
Most importantly, the entire product is designed to be used without a programming background, knowledge that is uncommon among traders and analysts. “If you are a quant, it isn’t as difficult to get data and clean it up. We are targeted toward non-programmers, who don’t have the same abilities. They may trade millions of dollars, but they don’t write code,” Zhen says. He hopes to maintain a focus on keeping the platform open to the widest number of people as possible.
Thinknum’s potential extends beyond getting a bunch of analysts on Wall Street to stop emailing spreadsheets back and forth. The founders hope to expand the company toward emerging markets where sophisticated financial tools are often lacking and analysts are disconnected from the global investment community.
While emerging markets have tanked in recent months due to uncertainty related to the fed’s bond buyback program and instability in Ukraine, Thailand, and Venezuela, there remains a deep market for better financial understanding of emerging market companies. Thinknum’s founders see better models as the first step to building a stronger investment climate.
“Creating communities of trust allows us to create insights into the market,” Zhen says. “If you are able to create a large number of models, and get an average of all of those models, that number is really, really powerful.”
He hopes the impact of the product isn’t just helping investors make better financial decisions, but also helping groups like farmers better navigate a volatile global trading environment.
“Let’s say you are a tomato farmer in India. How would you price your tomatoes? How do you know what the trends are? With the explosion of Internet access, we believe we will provide a lot of value in connecting these people to global trends in prices. Right now, it may seem like a sophisticated tool for financial analysts, but our grand vision is to provide this information and insight to all kinds of different types of people.”
For a company only a couple of months into production, there is obviously much to do. The sharing mechanisms can be a bit clunky, and the user interface has definitely been bootstrapped by a pair of former investment professionals. But the company already has several paying customers, indicating at least some desire on the part of the industry to find a better toolset. The company hopes that allowing charts to be embeddable in websites will give it wider distribution, and further improvement to its user experience should broaden the site’s appeal.
As investment banking undergoes a number of changes (like requiring that entry-level analysts take Saturdays off!), it’s time to look into the technical infrastructure of financial analysis and improve upon tools first developed in the 1980s. Along the way, we have the opportunity to open up finance from a handful of major cities to anyone with an Internet connection.
Image from Shutterstock/isak55 (image cropped)