The challenge
The challenge
A global sales digitization program needed to support more than 90 processes without tying the underlying business data to a single application interface.
Approach
The business workflow came first
The engagement covered the digitization and automation of more than 90 global sales processes. Power Apps provided the business-facing experience, with automated approvals and process state management.
Reminders and escalations formed part of the workflow design. The program connected the application experience to the progression of each business process, rather than treating the interface as an isolated form.
Separate the interface from the system of record
A central architecture decision was to separate the Power Apps user experience from the enterprise data layer. Dataverse, formerly Common Data Service, was not made the primary system of record.
The shared data foundation was built on AWS. This kept the same sales-process data available for reuse across applications and analytics, with a foundation for future integrations.
The boundary mattered because a business interface and an enterprise data model have different responsibilities. The application supported the task in front of the user. The shared data layer supported uses beyond that individual application.
A practical cross-cloud delivery
Microsoft Power Apps and AWS served different parts of the same program. The architecture combined the business workflow experience with a separate, reusable data foundation.
The result was a sales digitization program with a shared enterprise data layer. More than 90 business processes could use the workflow experience while the data foundation supported reuse across the wider application landscape.
Inside the implementation
Architecture overview
Power Apps handled the user experience. Workflow controls managed process progression. A separate AWS data foundation made the same business data available beyond an individual application.
Sales workflow and shared data
- 01
Business experience
The user-facing layer for global sales processes.
- Microsoft Power Apps
- Business-facing workflows
- 02
Workflow controls
The progression of work through its business states.
- Approvals
- Reminders and escalations
- Process state management
- 03
Enterprise data layer
The shared data foundation was built separately from the application experience.
- AWS data foundation
- Enterprise sales-process data
- 04
Wider enterprise use
The same information could support more than one business application.
- Other applications
- Analytics
- Future integrations
Application and data boundary
Power Apps was the experience layer. Dataverse was not made the primary system of record.
Process ownership
Approvals and escalations belonged to the business workflow, alongside its state management.
Key design decisions
Keep data useful beyond one interface
Separating the enterprise data layer from Power Apps allowed the same process information to support other applications and analytics. Reuse was part of the architecture rather than a later export exercise.
Model the work as a process
The delivery combined business-facing workflows with state management. An approval, reminder, or escalation was part of the progression of work, rather than an isolated notification feature.
Use each platform for its assigned responsibility
Power Apps provided the business interface while AWS provided the shared data foundation. The program therefore connected two platform responsibilities within one sales digitization effort.
How the workflow fits together
Business activity enters the workflow
The business-facing Power Apps experience is the entry point for a sales process.
The process advances through its controls
Approval steps, reminders, and escalations operate alongside the process state.
Process information has a shared data home
The enterprise data foundation on AWS separates reusable business information from the user-interface layer.
Other consumers can use the same foundation
Applications and analytics can reuse the sales-process data, with future integrations supported by the same architectural separation.
Questions for a similar implementation
Use these review points when you assess this architecture for your own environment.
- Where is the authoritative state of each business process?
- Which data belongs to a specific screen, and which belongs in the shared enterprise model?
- How will a change to a process affect other applications that use its data?
Results
The outcome
More than 90 global sales processes were digitized and automated. Their shared data foundation on AWS supported reuse beyond a single Power Apps workflow.
Stack
Technology used
- AWS data foundation
- Microsoft Power Apps
- Enterprise data integration
- Workflow automation
Client and delivery-partner names are withheld.