Description

There is a need to test and experiment against the existing model in revenue management.
Revenue management team has cultivated various pricing principle used in manual execution for pricing.

Problem

ATLAS' decisions have little explainability.
We need a model type that is highly configurable to the retrieved knowledge by revenue management.

Deploy MARS, a revenue management model/player that is based on set rules. issue
Enhancement towards the received ADR as input as well as seasonality issue

[x] Achieve at least 20% in-month revenue contribution by Mars. Constraint to revenue from 2024-04-14 after deployment.
- Achieved 21.82% of in-month revenue for the stated period.

Overall Point: 5.6

Functional Complexity: 0.9

Integrating a new revenue management model involves aligning it with existing business logic, which is moderately complex.

Technical Complexity: 1.1

Developing a configurable model like MARS requires advanced technical understanding of revenue management systems and configurability.

UI/UX Complexity: 0.5

This task might not have major UI changes but might involve dashboard updates for displaying revenue contributions from MARS.

Data Manipulation: 0.8

Data manipulation will be significant for experimenting and deploying the new model but will be based on established principles.

Testing: 0.3

The need for testing is inherent, but since this model is an enhancement with measurable metrics, there's less uncertainty in the expected outcomes.

Dependencies: 0.4

There may be dependencies on external data sources and internal systems, but given the clear roadmap, these are predictable.

Risk and Uncertainty: 0.6

Though a new model introduces risks, the clear SLA and existing principles reduce uncertainty about its performance.

User Impact: 1

A new revenue player like MARS designed to meet a specific revenue target will have a high impact on business outcomes and user success metrics.