The Three Cornerstones of a Smart Business

Adaptable Products.

Algorithms cannot iterate without the products— the online consumer interface that delivers customer experience directly while gathering consumer feedback to adjust algorithm models. Google’s search bar is a classic example of product design. Customers enter keywords into a simple bar and then immediately see the product—the search result page. Google puts tremendous resources into designing this product so that customers can find what they want in less time and with less effort.

Datafication and Live Data.

The starting point for the smart system that underlies Mybanks microloans is capturing business problems in digital form. How should offline business activities be rendered into data? Beyond simple operations and sales metrics, which present numbers that are easily digitized, other valuable questions can be answered through the use of more indirect data points. For example, to gauge how committed and active a seller is, Alibaba’s earliest microloan product looked at how many seconds a seller takes to respond to customer queries and comments on products and services. Datification is not easy or cheap, but it is the most important step in data intelligence.

Machine Learning: Iterating Algorithms.

Data can only create value when it is processed through a machinelearning engine. For a business, algorithms must make explicit the underlying product logic or market dynamics that the business is trying to optimize. Besides constantly refining themselves, algorithmic engines can also test various scenarios to improve business performance. For example, Mybank’s data scientists embed an experiment into the lending interface, giving different groups of sellers different interest rates and measuring real-time response rates. Each time the model gives a seller a particular rate, the platform records the seller’s reactions in real time. This data in turn is used as input for the lending model to adjust its parameters in a constant stream of calibration.