Accelerate Negotiation Guidance addresses 3 major personas Pricing Managers, Pricing Scientists and Configuration Engineers. In this section we will address just the Pricing Managers and Pricing Scientists.

Negotiation Guidance Target Audience

What is a Science Brick

Open

From science to user

In short, a science brick is what we consider at Pricefx any building block contributing to the Price Optimization features, such as:

• Product Similarity 

• Clustering 

• Multifactor Elasticity 

• RHS Demand Elasticity

• Forecast 

• Attribute binning 

• … and others

Science bricks are utilized for segmentation, elasticity, and clustering tasks performed by data scientists on a daily basis. Pricefx transforms these bricks into productized features by combining them with enablers (BE, Python, OE) provided by the engineering team. In simpler terms, we create and assemble these bricks into enablers, develop new science prototypes, test them using selected datasets or proof of concepts, and ultimately generate optimization accelerators. This process, known as optimization use cases, enables us to efficiently deliver pre-packaged solutions to users and customers in need. The pricing science team collaborates with the accelerator team and leverages the platform manager application to facilitate this process.

To support the creation of optimization use cases and accelerators, we adopt a component-oriented vision where the bricks are assembled. However, a proper framework is required to run complex processes, often involving advanced machine learning algorithms or AI. These processes entail data preparation, data mapping, model training, and utilization in other models for optimization and prediction purposes. The framework eliminates the need to constantly adapt the UI and UX for each new use case, providing a seamless experience. Pricefx can deliver use cases based on the accelerator concept, making it easy to plug and deliver bricks with just a few clicks to any partition for any customer.

Sample Mini Use Case with Science Bricks - Multifactor Elasticity

As an example, the multifactor elasticity science brick, a component of this accelerator, utilizes transactional data, product information, and customer data.

By incorporating event-specific information such as Black Friday or a targeted week when an even is scheduled, it becomes an input for optimization and multifactor elasticity. Using the Python engine, this feature enables forecasting based on product attributes and similarity scoring. In turn, it facilitates the creation of new product clusters.

This is particularly useful for product recommendations, where similar products with better margins or lower prices can be suggested.