Our Approach

How a system gets from a requirement to something that runs

Six stages, in the same order every time. The sequence exists because the expensive failures in this work happen when integration and validation are treated as things to sort out later.

Methodology

Six stages

Each stage produces something the next one depends on. We do not start building until the architecture is agreed, and we do not deploy until validation has passed.

  1. Discover and assess

    Understand the operating environment, the existing systems, and what the people who will use the result actually need.

  2. Define architecture

    Decide the shape of the system — layers, platforms, data flows and integration points — and record why.

  3. Design the solution

    Turn the architecture into a specification concrete enough to build and to procure against.

  4. Integrate platforms and systems

    Connect platforms, devices and existing software so they operate as one system rather than several.

  5. Test and validate

    Verify behaviour against the requirement, in conditions resembling the ones it will run in.

  6. Deploy, support, and improve

    Put the system into service, support it in operation, and extend it as requirements change.

Phase one

Understand

Covers stage 1 — Discover and assess

We start on site rather than in a document. What equipment exists, what condition it is in, what connectivity is actually available, and which systems already hold the data that matters.

The most useful part of this phase is usually finding out what the requirement omits. A brief asking for vehicle tracking often turns out to be a request for dispatch visibility, which is a different system.

What happens

  • Site and environment assessment
  • Review of existing systems and the data they hold
  • Interviews with the teams who will operate the result
  • Constraint capture — connectivity, power, access, downtime windows

What you get

  • A written statement of the current state
  • A requirement clarified with the people who raised it
  • A list of constraints and open questions, with owners

Phase two

Design

Covers stages 2 and 3 — Define architecture, then Design the solution

Architecture first, then specification. The architecture decides the layers and where responsibility sits between them; the specification makes it concrete enough to build and to buy against.

Both are written down with reasoning attached. A design decision without a recorded rationale gets reversed by the next engineer who does not know why it was made.

What happens

  • Layer and component architecture
  • Platform selection against the assessed constraints
  • Integration point and data flow definition
  • Specification detailed enough to procure and build from

What you get

  • An architecture document with decisions and rationale
  • A build-ready specification
  • A procurement list tied to the design rather than to a vendor

Phase three

Integrate

Covers stage 4 — Integrate platforms and systems

This is the phase most often underestimated. Connecting a new platform to systems that were never designed to share data is where deployments overrun, and it is why integration is scoped and priced up front rather than handled as variations.

We build against the customer's actual systems, not a documented ideal of them. Interfaces are tested with real data before anything depends on them.

What happens

  • Platform configuration and commissioning
  • Interface development against existing systems
  • Device and gateway provisioning where applicable
  • Data mapping and normalisation across sources

What you get

  • Working interfaces tested with real data
  • A configured platform connected to the existing estate
  • Integration documentation for the customer's team

Phase four

Validate

Covers stage 5 — Test and validate

Validation checks behaviour against the requirement captured in phase one, not against the specification alone. A system can meet its specification and still fail to answer the question the customer asked.

Testing happens in conditions resembling the operating environment, including the degraded ones — intermittent connectivity, partial data, equipment offline.

What happens

  • Functional verification against the original requirement
  • Integration testing across connected systems
  • Behaviour under degraded connectivity and partial data
  • Access control and permission verification
  • Review with the teams who will operate the system

What you get

  • Test results traceable to requirements
  • A defect list with severity and resolution status
  • Sign-off from the operating teams, not only the project sponsor

Phase five

Deploy

Covers the first part of stage 6 — Deploy

Deployment is planned around the customer's operating constraints, not ours. Infrastructure organisations rarely have a convenient window, so cutover is sequenced to keep existing services running.

Handover includes the knowledge, not just the credentials. Delivery is complete when the customer's team can operate and explain the system.

What happens

  • Cutover planning around available downtime windows
  • Staged rollout where the environment allows it
  • Operator training on the views each team will use
  • Documentation handover

What you get

  • A system in service
  • Trained operators with documented procedures
  • An agreed support and escalation path

Phase six

Operate and support

Covers the rest of stage 6 — Support, and improve

The system's operating life is longer than its delivery, and most of its value is realised there. We stay with it — resolving field issues, adapting to changed requirements, and keeping it serviceable as teams and equipment change.

Improvement is deliberate rather than continuous churn. Changes are made because an operational need has been identified, not to keep a release cadence.

What happens

  • Monitoring and field issue resolution
  • Configuration changes as operations evolve
  • Platform updates coordinated with the vendor roadmap
  • Periodic review of security posture and access

What you get

  • A supported system with a known escalation path
  • A record of changes and why they were made
  • Extension work identified from operational need

Throughout

Security and quality across every phase

Neither is a stage. Both are applied within all six.

Treating security as a phase produces a system that is protected at the perimeter and nowhere else. Treating quality as a phase produces testing that finds design problems far too late to fix cheaply.

So both run throughout. Security requirements are captured in discovery, decided in architecture, built during integration, and verified in validation. Quality practice means decisions are written down, work is reviewed by another engineer, and defects are tracked to closure rather than to the end of a sprint.

  • Recorded decisions

    Architecture and platform decisions are documented with their reasoning, so they can be reviewed rather than rediscovered.

  • Peer review

    Design and integration work is reviewed by an engineer who did not produce it, before it reaches the customer's environment.

  • Security at each layer

    Device credentials, network segmentation, transport encryption and platform access are specified alongside the system, then verified in validation.

  • Traceable testing

    Test results map back to the requirement they verify, so coverage is visible rather than asserted.

  • Defect tracking to closure

    Defects carry a severity and a resolution status, and are closed rather than deferred silently.

On certifications

These are the practices we work to. ZouhThings does not currently claim any formal quality or security certification, and nothing on this page should be read as a statement of accreditation or standards compliance.

Delivery model

Partnership-led delivery

We work with the customer's team and with our technology partners, rather than delivering to a specification from a distance.

Two relationships shape how a project runs. The first is with the customer's own engineers, who will operate the system after handover and therefore need to help shape it before it is built. The second is with the platform vendors, whose products we deploy and whose roadmaps the system will follow.

That model has a practical consequence: we are not the only party who can maintain what we build. Established platforms with vendor support, plus a customer team that understands the system, means the organisation is not dependent on us to keep it running.

  • Customer engineers involved during delivery, not briefed at handover
  • Established international platforms rather than bespoke code where a product will do
  • Vendor-maintained roadmaps, so the platform develops independently of us
  • Implementation, integration and support handled locally in Damascus
  • Documentation written so the system can be maintained without us

Get in touch

Discuss how we would approach your project

Tell us about the environment and the constraints, and we will explain what discovery would involve.