Revenue growth & profitability
Data-driven pricing strategies and competitor price simulation to optimise revisions and capture market opportunities.
Spare Parts Price Information Management System
An analytics-driven pricing platform that replaces spreadsheet-era spare parts pricing with rule-based opportunity detection, ML price suggestions, and revenue/profit simulation — so every revision is deliberate, measurable, and fast.
- 2×/yr
- Current revision cadence
- → continuous, trigger-based
- 0
- Cost inputs in today's price
- → cost + competitor + demand
- 3
- Workstreams
- Evaluate · Analyse · Execute
- 12 mo
- Forecast horizon
- Parts demand forecasting
01 — Problem statement
Pricing runs on spreadsheets, memory and a twice-a-year calendar
Nissan's current manual spare parts pricing process relies on outdated legacy systems and ad hoc revisions. It lacks data-driven insight, so profit opportunities are missed and the team absorbs the inefficiency. An analytics-based pricing management system enables strategic revisions, revenue simulations and dynamic adjustments driven by cost, competitor data and market conditions.
Cost of doing nothing
Every month between revisions is a month of prices that no longer reflect cost, competitor position or demand.
Excel
System of record
Manual
Validation
Historic
Reference basis
Prices are adjusted from historical references in an Excel-based legacy tool, then uploaded to the publication system with no automated validation and no measured outcome.
02 — Business objectives
Six outcomes the system is accountable for
Process automation & efficiency
RPA for value-driver extraction, cutting manual effort and easing limited manpower capacity.
Advanced decision-making
Real-time visualisation and simulation for pricing decisions — better accuracy, responsiveness and planning.
Independent & scalable system
Deployed outside the legacy framework for flexibility, autonomy and minimal dependency on outdated stacks.
Centralised data management
A robust data platform delivering pricing accuracy, consistency and accessibility across all pricing functions.
AI/ML-driven optimisation
A complete price revision history that feeds ML pricing models for continuous, smarter adjustments.
03 — Business flow
From manual sheets to an analytics loop
The future state uses an analytics-driven approach: profit and revenue simulation on parts forecasting data, a dynamic pricing model built on cost, revision history, competitor prices and category segmentation, and automated revision triggers driven by predefined scenarios and business rules.
Inputs
- Centralised Pricing Data Repository
- Sales Data
- 12-Month Parts Forecast
Hosted on AWS Cloud
STEP 01
Rule-based opportunity evaluation
User-defined static and statistical rules surface eligible parts.
STEP 02
Price calculation using ML models
Suggested prices from cost, revision history, competitor and segment.
STEP 03
Revenue & profit simulation
Forecast-driven what-if on margin, revenue and volume uplift.
STEP 04
Price finalisation & execution
Approved revisions pushed to the publication system.
Rule / Forecast / Price EnginesSimulated & approved revisionPrice Publication PlatformCycle time: on-trigger · Validation: simulated before execution
04 — Solution architecture
A View Layer users drive, a Model Layer that does the maths
View Layer
Identify parts eligible for revision and the suggested price at a point in time.
- Rule-based system: users create and maintain rules through the UI
- Static rule components (filters, thresholds, last-revision age)
- Dynamic components — statistical and ML-based price logic
- Rules run as scheduled batch jobs or Just-In-Time
Model Layer
Rule Engine
Identifies the parts eligible for price revision from user-defined and statistical rules.
Forecast Engine
Generates 12-month part-level forecasts to quantify future demand.
Price Engine
Calculates new price suggestions for revisions and for newly introduced parts.
Sales data and the centralised pricing repository feed the engines; finalised prices flow out to the Price Publication Platform.
05 — Product walkthrough
The View Layer, live — change the inputs and watch the numbers move
All four screens are clickableEvaluate — Rule BuilderInteractive prototype
Create new rule
Identify parts
Price revision logic
Price Rule Preview
4 parts eligible| Desc | Part Code | Part # | Last Rev. | Current | New Price |
|---|---|---|---|---|---|
| A00 | PRT-1187 | A00-1187-FP | 14 mo | ¥12,400 | ¥13,640 +10% |
| A00 | PRT-2043 | A00-2043-BR | 26 mo | ¥8,600 | ¥9,460 +10% |
| B12 | PRT-4521 | B12-4521-SP | 31 mo | ¥15,800 | ¥17,380 +10% |
| C07 | PRT-5090 | C07-5090-RD | 19 mo | ¥22,400 | ¥24,640 +10% |
Figures are illustrative sample data used to demonstrate the interaction model, not actual Nissan pricing.
06 — Impact & roadmap
What changes once the system is live
Strategic
Price revisions
Selection driven by rules, cost and forecast instead of flat percentage sweeps on EOP and low-margin parts.
Continuous
Revision cadence
Automated triggers fire on predefined scenarios and business rules rather than a twice-a-year calendar.
Simulated
Every decision
Revenue, margin and ROI impact are quantified before a price ever reaches the publication platform.
Auditable
Revision history
A complete history of revisions and outcomes becomes the training set for AI/ML pricing models.
- 1
Phase 1
Centralised pricing data platform on AWS
Repository, sales data ingestion, RPA value-driver extraction.
- 2
Phase 2
Rule Engine + Evaluate workstream
UI rule builder, batch and JIT execution, price rule preview.
- 3
Phase 3
Forecast & Price Engines + Analyse
12-month forecasting, ML price suggestion, financial / competitor / ROI analysis.
- 4
Phase 4
Execute + downstream integration
Approval workflow, publication integration, ML feedback loop on outcomes.