We’ve been hearing about Data Science as a Service (DSaaS). It can be a good option based on your company’s needs combined with the shortage of talented, experienced Data Scientists.
Serial Metrics is one such company offering DSaaS. They are a Data Science company with a team of PhDs from top-tier universities that discovers value in data and enables data driven opportunities. I thought it would be helpful to our readers to see what kinds of problems they are solving via the DSaaS approach.
Here are 4 Case Studies
AAA Membership Services Cost Forecast
Serial Metrics custom-built a system dynamics simulator to forecast emergency roadside costs for AAA Membership Services. The challenge: several, highly variable, and difficult to predict, macro-economic drivers, such as the unemployment rate, fuel prices, and even weather patterns determine annual operational costs. In addition, usage, member adoption rate, member attrition rate, word of mouth (WOM), marketing efficacy, etc., all translate to a complex cost forecast.
Eventbrite Dynamic Pricing
Serial Metrics designed a novel, patent pending, pricing application that estimates real-time, instantaneous demand, using the velocity of sales: quantity sold per unit of time (Qs/t). In addition, price elasticity of demand [(%∆Qs/%∆P) * P/Qs] is approximated in real-time to determine the revenue maximizing price. This method of pricing is of particular value to organizations that sell perishable commodities such as concert tickets.
WayPoint Real Estate Group – Optimizing Portfolio Purchases
Waypoint Real Estate Group LLC, buys, fixes, and rents foreclosed homes, nationwide, and is among the largest investors in the U.S. real estate market, currently owning upwards of 10,000 properties. Their business model presumes buying a large volume of inexpensive properties, while minimizing any associated inspection and renovation costs, then renting the properties at a rate that yields a steady profit. To determine which homes Waypoint should purchase, and then renovate, usually requires “walkthrough” property inspections, where – for example – a property inspector evaluates worn kitchen cabinets or missing roof tiles. The process is time consuming and, itself, costly; more importantly: physically, inspecting properties is an unscalable process for a company whose business model requires time-sensitive, bulk home purchases.
To scale operations, Waypoint relies upon a comprehensive software application to identify and evaluate which properties, based on expected renovation costs, would prove profitable to purchase, and subsequently lease. The basis of such a software application requires a set of machine learning algorithms to quickly determine – without physically inspecting properties – which homes are worth purchasing for profitable returns. READ MORE: (PDF Version)
Sycle Lead Routing Algorithm
Serial Metrics designed and implemented an innovative, machine-based, lead nurturing and routing application using a proprietary machine-learning algorithm to decide email campaign target population (who receives the message) and message sequence (what message they receive).
The challenge: prospective customers (sales leads) avoid responding to direct marketing campaigns when the marketing promotion does not align with immediate customer needs. Meanwhile, sales leads that respond to marketing promotions already have some desire to purchase. Identifying the segment of sales leads that are “ready to buy” and those that are not is key. Further important is the ability to determine when to connect with the sales leads to ensure the maximum purchase likelihood. Traditional lead nurturing processes periodically correspond with sales leads, imprinting brand awareness, and cultivating interest so at the ‘right’ time a sales lead converts to purchase. READ MORE (PDF Version)
If you are interested in learning more.