Machine learning enables your business to make better decisions using data.

Our goal with the Vidora Decisioning SDK is to empower teams to take full advantage of both their real-time and batch data to make the best decisions for every user. The Decisioning SDK is ideal for businesses creating dynamic decisioning experiences like next-best-action and next-best-offer experiences. The Decisioning SDK is designed for scale and low-latency decisions, so that businesses can use it within product experiences (which require low-latency responses) across millions of unique users.

Building a machine learning-powered decisioning experience requires that teams have access to more than just an inference engine to make predictions. Additional areas to consider include –

  • Model types which are available (for instance is prescriptive modeling available?)
  • Ability to add custom business logic to the decisions, such as the business value of a particular conversion
  • Ability to add new interventions and offers, and have those automatically incorporated in the decisioning
  • Easy visualization of the performance gains over a control experiment
  • Flexibility around the data and features used to build the decisioning

Examples of features available through Cortex and the Decisioning SDK.

Screen-shot of building a real-time pipeline within Cortex. This pipeline determines whether to show a hard paywall or registration paywall at a breach event for a user in order to maximize total subscription conversions.

An Example of Next-Best-Offer – Discounting a Subscription

Let’s take an example – offering a discount to users to join a subscription service on a product website. Let’s consider that a business is able to offer a $10 discount to users. The question naturally arises, which users should receive the discount? This is a great use case for the Decisioning SDK.

Here are a few configurations you will have access to with the Decisioning SDK to maximize your business goals –

  • Using Prescriptive Modeling – You will have the option of using prescriptive uplift modeling to target the users whose likelihood of subscribing will increase the most if offered the coupon. Prescriptive modeling will save your business money by not targeting users already very likely to subscribe. The Decisioning SDK will take care of the explore-exploit testing automatically which is required for prescriptive modeling.
  • Adding New Offers – The SDK will also allow you to easily add a new discount, let’s say $20, to the mix. It will seamlessly integrate the new offer to learn which users should be receiving it. You also have the ability to assign business rules like cost to each discount to ensure that every user is receiving the coupon which maximizes total business value.
  • Visualizing New Revenue – Finally, Cortex allows you to visualize the increased revenue obtained from using dynamic decisioning over a control group of users whose targeting decisions are random. This enables you to demonstrate the value of dynamic decisioning to your entire business. The SDK allows you to easily set and modify the size of your control group.

Wrapping it Up

We’ve built a broad set of functionality into the Decisioning SDK and we are engaging with our customers and partners on an ongoing basis to make it easier for them to build high-performing automated decisions throughout their business.

If you are interested in learning more about the Decisioning SDK or providing additional feedback, we would love to hear from you. Feel free to take a look at our online documentation and contact us directly at

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.