Strategy & Architecture
Solution architecture, technical discovery, system assessment and roadmap work -- the decisions made before a build starts, which are the ones that cost the most to reverse.
Problem framing
What tends to bring people here
A decision has to be made without a basis for making it
Build or buy, replace or extend, one platform or several. The choice is consequential, and the people asked to make it do not have comparable information about the options.
Vendor proposals cannot be compared
Each responds to a different reading of the requirement, so the differences in price reflect differences in scope that nobody has made explicit.
A programme has stalled and the cause is unclear
Delivery has slowed and the explanations conflict. What is needed is an independent read of the technical situation, not another status report.
Systems were bought one at a time
Each procurement was defensible on its own, and together they have produced an estate with overlapping capability and no integration story.
The requirement is not yet well enough understood to specify
There is a clear problem and no clear solution, and writing a specification now would lock in the wrong assumptions.
What we deliver
Concretely, what you receive
Technical discovery report
Current state, constraints, options with their trade-offs, and a recommendation with the reasoning left visible so it can be challenged.
Solution architecture
Target architecture, integration model, data flows, security model and the sequence to get there from where things stand.
System assessment
Independent review of an existing system against maintainability, security, performance, integration readiness and total cost.
Technology selection framework
Weighted criteria and a comparison an organisation can defend in procurement, rather than a preference presented as analysis.
Digital transformation roadmap
Sequenced work with dependencies, decision points and the capabilities the organisation needs to build alongside the systems.
Specification for procurement
A requirement written so that responses are comparable and the resulting contract is enforceable.
Typical engagements
Shapes this work takes
- architecture review
Review before committing
Independent assessment of a system, a proposal or an architecture, with a written recommendation.
- new platform
Discovery ahead of a build
Establish what should be built, in what order, and what must be decided before anyone writes code.
- modernisation
Plan a route out of a legacy estate
A sequenced modernisation path that keeps the organisation running throughout, rather than a replacement programme with a single cutover.
Engineering approach
How we go about it
Methods
Options with trade-offs, not a single recommendation
Present the realistic options and what each costs and forecloses. A recommendation without visible alternatives cannot be evaluated by the person accountable for it.
Constraints first
Budget, timeline, existing systems, procurement rules, connectivity and the capacity of the team who will operate it. Most architectures fail on constraints rather than on design.
Decisions recorded with their reasoning
Architecture decision records, so a future team can tell whether a choice still holds or was made against conditions that have changed.
Standards
Reversibility valued explicitly
Where two options are close, prefer the one that is cheaper to undo. The estimate of which is better is less reliable than the estimate of which is easier to escape.
Independence stated
Where we would be a candidate to build what we are recommending, that is declared in the report rather than left for the reader to notice.
Patterns
Strangler-style modernisation
Route capability away from a legacy system incrementally, so value arrives during the programme rather than at the end of it.
Boundaries drawn on ownership
Service boundaries follow who owns a process, because a boundary that cuts across ownership generates coordination cost forever.
Deliberate technology floor
Choose the least novel technology that meets the requirement, so the organisation can hire for it and support it.
Platforms we deliver on
Azure
Kubernetes
Terraform
Facing a decision that is expensive to reverse?
Tell us what is being decided and what constrains it. We will tell you what we would need to look at to give you a defensible answer.