From business question
to trusted data.
A customer strategy is only useful when the underlying data can support it. The technical challenge is to connect the right sources, model the right relationships and make the resulting information reliable enough to drive decisions.
Business Model
Define the customer model
Establish what a customer, relationship, product, transaction and opportunity actually mean within the business.
The important questions
Before choosing technology, establish the business definitions and relationships that the technology needs to represent.
- What constitutes a customer?
- How are customers related to entities and accounts?
- What represents a relationship?
- What does “value” mean?
Data Sources
Find and connect the evidence
Identify the systems and external sources containing the information needed to create the customer view.
Typical sources
The customer rarely exists in one system. The architecture needs to accommodate the reality of fragmented operational data.
- CRM and customer platforms
- ERP and finance systems
- Sales and marketing platforms
- Operational and service data
- External market and reference data
Integration
Bring the data together
Create reliable pipelines and integration patterns that move data from operational systems into a usable analytical environment.
Architecture decisions
Integration should be driven by business requirements rather than technology fashion.
- Batch vs near-real-time requirements
- APIs and system integration
- ETL / ELT patterns
- Cloud and on-premise environments
- Scalability and maintainability
Data Model
Build the customer view
Resolve identities, relationships and business definitions into a model that can answer the questions the organisation actually has.
From records to relationships
The goal is not simply to create another database. It is to create a coherent representation of the customer and their economic relationship with the business.
- Customer identity resolution
- Entity and relationship modelling
- Master and reference data
- Historical vs current state
- Customer 360° views
Quality & Trust
Make the information trustworthy
Data quality is not an abstract technical exercise. It determines whether the resulting commercial decisions can be trusted.
Trust requires evidence
Define the controls and lineage needed to understand where information came from and whether it is fit for purpose.
- Data quality rules
- Completeness and accuracy
- Data lineage
- Ownership and accountability
- Exception management
Insight & Delivery
Turn the model into decisions
Deliver the customer intelligence through reporting, analytics, workflows and decision-support tools that people can actually use.
The end point
The architecture succeeds when the right people can access the right information at the point where a business decision is made.
- Management information
- Self-service analytics
- Customer segmentation
- Predictive and analytical models
- Operational decision support
Want to connect the technical architecture back to the commercial case?