For any company, implementing machine learning is a massive challenge that can take up hundreds of millions of dollars and take years to fully realize. One of the ways to reduce those costs is to minimize deployment time.

Cortex makes ML accessible to non-technical stakeholders, allowing anyone at your organization to build ML pipelines that make predictions using your business’s data. Cortex automates the entire ML pipeline from raw data to prediction generation. In doing so, Cortex helps businesses implement machine learning in hours, not years.

This case study walks through how Cortex helped one of our real estate partners partners implement machine learning to solve a variety of problems.

The Problem

Our partner is a large global real estate business with a significant and growing digital presence. Real estate selling and purchase behavior increasingly migrates online. Our partner wants to ensure they maintain a unique voice in this crowded real estate landscape.

Our partner wanted to predict user behavior on both the buying and selling side of the transaction. However, they struggled to implement an internal ML solution at scale, so asked Vidora to help speed up the process with Cortex.

They wanted to deploy ML to address use cases to predict which users were most likely to be:

  • First home buyers
  • Users that engage with an agent on the platform via SMS
  • Home sellers

The Cortex Solution

Cortex offers a solution to this challenge. Cortex takes in raw, unprocessed data from a variety of sources and centralizes all of your business’s data in one place. As a general rule, the more data available, the better Cortex will perform. But don’t worry about throwing as much data as possible at Cortex. Cortex will automatically determine the right data to use when building pipelines by relying on various automated feature engineering techniques.

In this case, Cortex collected data from over three years of history. Our partner also was able to use Cortex’s easy integration tools to establish the infrastructure necessary to collect this user behavioral data on an ongoing basis in real time. Here is a breakdown of the data collected by Cortex:

deployment

Here are some of the pipelines our partner built in Cortex:

First-Time Home Buyers:

deployment

Which Users Would Engage with an Agent

deployment

Which Users Intended to Sell:

The Results

Using Cortex to implement machine learning across their business, our partner was able to reduce machine learning deployment time by over 25x. Typical pipeline configurations included:

  • 6 billion+ raw events
  • 500+ engineered features
  • 20+ ML pipelines

What had originally taken our partner close to a year to build was handled by Cortex within a couple of weeks. As a result, our partner was now able to predict user behavior in a number of areas in their business. This resulted in a faster and more powerful ML solution than had previously been possible.

Summary

Machine Learning has revolutionized the way companies think about optimizing product strategy for their customers. Your business could be leaving millions of dollars on the table every year without technology like Cortex, which enables anyone – even non-technical users – to build complex ML pipelines with just a few clicks and save months of development time.

Want to know more? Contact us here, or email us at info@vidora.com!

Want to Learn More?


Schedule a demo and talk to a product specialist about how Vidora’s machine learning pipelines can speed up your ML deployment and ultimately save you money.