Our Gen IV AI will blow your neural network

PROS created the Profit & Revenue Optimization Software category, and we remain the undisputed leader in the space. With the launch of the PROS Platform, we now have a flexible and modular platform which future proofs customers who want to build organic margin by optimizing costs and revenue in existing operations.


Gen IV


Neural Network

Use the latest deep neural network to dramatically improve willingness-to-pay prediction without sacrificing interpretability through explainable AI.


Price Prediction: Completely eliminate segmentation by turning it into a prediction problem using a deep neural network that use all available data and attributes.

Price Optimization: Provides customer-specific, market-aware, and win-rate based pricing.


  • Significantly improved prediction accuracy from deep neural network
  • Eliminate data sparsity by eliminating segmentation
  • Ability to use categorical features with huge number of values
  • Ability to use external data sources (e.g. market indices, competitor data)
  • Ability to use loss information
  • Automatic modeling of seasonality and trends
  • Target price is determined via profit/revenue optimization instead of heuristics and business rules
  • Peers are predicted dynamically based on features derived from transaction, customer, and product.
  • Explainable AI provides transparency to the neural network
  • Extensible to non-negotiated guidance



Dynamic Pricing

Major move to user-driven workflows so that the science moves out of the backroom and into the application itself.


Segmentation: Online SKU-centric symmetric spanning tree that uses dynamic attribute roll-up.

Price Recommendation: Customer-level analysis with benchmarking and guidance against peers that respects user-controlled price change aggressiveness.


  • Product centric
  • Gradual correction of underperformers
  • Easy in-app user validation
  • Simulation capabilities including ROI estimation
  • Full user control of all aspects of segmentation and pricing guidance
  • Further improved sparsity handling with dynamic rolled up and re-segmentation
  • Meaningful floor and expert price to aid negotiation


  • Limitation on number of features and feature values
  • Inability to use external data sources to augment volatile historical data
  • Inability to use loss data directly
  • Fixed pre-selected attributes for segmentation

Gen II


Decision Tree

AI advances continue as we move to the use of data-science driven segmentation to mitigate the Cartesian data sparsity problem.


Segmentation: Supervised machine learning of asymetric binary tree model (based on CART) with advanced model fit statistic that uses Bayesian Information Criteria to prune the decision tree.

Price Recommendation: No change.


  • Improves prediction offered by CART’s flexibility
  • Dynamic attributes
  • Recency weighted transactions
  • Improved data sparsity handling (for segments with insufficient transactions or customer diversity)


  • Difficult to visualize which hurts adoption
  • Manual rules for  achievability and sensibility of target pricing
  • Offline attribute selection and segmentation

Gen I


Cartesian Model

AI begins in its basic form. Customer-specific willingness-to-pay price guidance to provide the right price to the customer.


Segmentation: Symmetric cartesian cross-product algorithm using Bayesian Information Criteria (BIC). Bucket feature values and create all possible combinations from all the buckets, then use BIC to iteratively select the final segmentation.

Price Recommendation: Percentile-based algorithm and broad peer groups.


  • Easy to understand and explain
  • Consistent peer groups
  • Placeholder segments
  • Customer specific pricing (takes into account of low and high performers)


  • Global and static attribute groupings
  • Rigid data structure leading to data sparsity
  • Manual rules to limit large price changes 
  • Offline attribute selection and segmentation
  • Lack of explanation for price recommendations in UI

Gen Zero



Foundational price management technology. Replaces Excel and integrates into back-office systems.


  • Greater efficiency through automation
  • Centralized platform for all pricing and cost information
  • Price transparency and visibility
  • Granular drill down (enables pricing at lower-level components, which roll up to the whole product)
  • Exception identification based on alerts and thresholds
  • Streamlined workflow
  • Ability to use leader-follower relationships on product portfolio 
  • Flexibility to build complex business rules


  • Hard to maintain business rules
  • Only rely on human decision
  • Relies on broad (business/marketing) customer segments
  • Matrix-based discounting structures