From complexity to clarity.
From requirements to real solutions.
Every engagement moves through the same seven stages, each led by the discipline it needs most. You can engage us for a single phase or for a complete project.
Stage 01 · Joint
Discover
Understand business operations, stakeholders, systems, limitations, pain points and objectives.
Typical outputs
- Stakeholder map
- Problem statement
- Success criteria
Ways to engage.
Engagements are scoped around the stages you need. Commercial terms are agreed per project.
Discovery & assessment
Stages 01–03. A clear problem statement, data and systems assessment, target architecture and a phased, costed roadmap. Useful on its own before procurement.
Build & integrate
Stages 03–06. We design, build, validate and deploy a defined solution, with documentation and handover.
Data & reporting improvement
Stages 02, 05 and 07 focused on data quality, reconciliation, governance and reporting.
Ongoing improvement
Stage 07 under an agreed support arrangement: monitoring, controlled enhancements, additional automation.
Two disciplines in every stage.
Switch viewpoints to see who contributes what — and why the combined design is stronger than either alone.
Example scenario: a fleet operator suspects fuel losses across remote sites
ENG · 01
Flow meter & PLC
Pulse acquisition and calibration at the dispenser.
DATA · 01
Define the measure
What exactly counts as one dispense, and in which unit?
Measurement definition
ENG · 02
Edge buffer
Store-and-forward so no dispense is lost offline.
DATA · 02
Source authority
Which record wins when meter, ERP and invoice disagree?
Source of truth
ENG · 03
MQTT transport
Secure, at-least-once delivery from site to server.
DATA · 03
Validation rules
Vehicle exists, volume ≤ tank capacity, no duplicate IDs.
Rules at ingest
ENG · 04
API & database
Idempotent ingest, history and audit tables.
DATA · 04
Reconciliation
Dispensed litres vs ERP issues vs supplier deliveries.
Matched records
ENG · 05
Dashboard & alerts
Live levels, alarms and data-age indicators.
DATA · 05
KPI model
Litres per km, variance %, exceptions by site.
Trusted KPIs
ENG · 06
Deploy & monitor
Docker services, TLS, health checks, backups.
DATA · 06
Governance
Owners, review cadence and exception workflow.
Accountability
Both partners, one design
Each engineering component exists because a business rule needs it, and each rule is enforced somewhere real. The result: every litre measured, matched to a vehicle and a transaction, with variances raised to an accountable owner.
- Rules are enforced at ingest, not discovered at month-end
- Reports carry their own data quality
- Exceptions have owners
- Engineering decisions trace to business value
Working principles.
- Design for real conditions
- Intermittent connectivity, remote sites, shift work and imperfect data are requirements, not edge cases.
- Make failure visible
- Stale data, failed rules and broken links should be obvious to the people relying on the system.
- People decide
- Automation and AI prepare and route work. Accountable people approve what matters.
- Validate before reporting
- Numbers are only presented once their source, rules and quality are known.
- Work with the platform
- ERPNext, Power BI and the Microsoft stack are extended the way they are designed to be extended.
- No invented outcomes
- We agree measurable targets during discovery instead of promising results in advance.
Let’s build something that works.
Whether you need a custom application, better visibility into your data, connected operational systems, or a practical automation strategy, we can help define the right solution.