Viva: Model-Based Business
by Benjamin Erichsen
Famous (or infamous, depending on who you ask) hedge fund manager Stevie Cohen wrote an informative piece that was recently published in the Wall Street Journal. He begins by discussing the software revolution that has swept the world over the past two decades. From there, he segues into what he sees as the inevitable rise of businesses built on models:
A model-driven business is something beyond a data-driven business. A data-driven business collects and analyzes data to help humans make better business decisions. A model-driven business creates a system built around continuously improving models that define the business. In a data-driven business, the data helps the business; in a model-driven business, the models are the business.
What Stevie describes is a self-perpetuating cycle of success that results when a model-driven business is firing on all cylinders. Data is the key ingredient for these models. All the artificial intelligence and machine-learning in the world aren’t of much use without the data needed to power such systems. This data is so important that Cohen posits that start-ups could find themselves forced into the position of purchasing incumbent companies for the data they posses if it can’t be obtained through other means.
Cohen goes on to explain that the model-based companies best positioned for success are those that possess “closed-loop” data systems. In a nutshell, these systems are ones where the business controls all the inputs and outputs throughout the customer’s decision-making process without relying on any third parties to complete the picture. Done right, these closed-loop data systems lead to a self-improving cycle:
- The models use these data systems to improve products
- Improved products attract more users
- More users bring more data for the model
- More data for the model means further improvement…
Viva is just such a company. Data-driven artificial intelligence is a core component of our business’s mechanics. Two examples of closed-loop data systems within Viva’s operations:
- When a customer purchases a Fractional Mortgage Share on Viva’s FMS Exchange, our model will take note of inputs such as a customer’s risk profile, amount of capital available to invest, geographic location and other data while still preserving that customer’s anonymity (the customer’s name isn’t a useful data point in any case). The outputs will take the form of the user’s decisions: FMS products that were viewed, those FMS products that are ultimately purchased, time spent browsing the Exchange, etc.
- When a borrower uses Viva’s platform to obtain a mortgage, the inputs will consist of the mortgage amount, borrower’s credit rating, value of the home being purchased, geographic location, interest rate desired by the borrower, etc. Outputs from this part of the business include whether the mortgage is successfully funded, the length of time needed to fund the mortgage, the interest rate obtained, the duration of the note, etc.
Viva’s models will make use of the data referenced in the above examples to improve the experience for both lenders and borrowers.
In example 1, the algorithm that determines which mortgages the FMS Exchange will display following a search will more accurately reflect the investor’s needs as the model learns from the data collected during all previous searches.
In example 2, Viva’s algorithms that will assist borrowers in choosing realistic loan parameters will get better and better at determining the parameters that will result in an expeditious funding of the mortgage as it learns from the experiences of other borrowers.
We at Viva are excited to bring the power of machine learning and artificial intelligence to bear on the global mortgage market as we level the playing field for borrowers and investors alike. Stevie Cohen says that models will run the world. That may be true but we are more excited about our models putting the world in homes.
Join us: www.vivanetwork.org
