Engagement models

Four commercial shapes, no ambiguity about who owns what.

The model is chosen for the risk profile of the work, not for how it suits us. Most programmes start with one shape and move to another after the first production release.

Fixed-scope delivery

Best for: A defined outcome with a date that cannot move.
Scope: Agreed in writing after discovery: deliverables, interfaces, acceptance criteria, exclusions.
Commercial: Fixed fee, milestone-based, with a change-control route for anything genuinely new.
Ownership: NAFSoft owns delivery; you own acceptance.
Reporting: Milestone progress, defect status, risk log.

Dedicated pod

Best for: Continuous product or platform work with a stable team.
Scope: Evolving backlog, reviewed monthly, prioritised by your product owner.
Commercial: Monthly rate for a named team, sized up or down inside the agreed notice period.
Ownership: Your product owner directs; our lead is accountable for quality and delivery.
Reporting: Monthly delivery report, velocity and quality measures, capacity forecast.

Time & materials

Best for: Discovery, feasibility work, or scope that is still being defined.
Scope: Work orders against an agreed rate card, with a written estimate of remaining effort.
Commercial: Hours consumed at the agreed rates, reported by workstream.
Ownership: Joint: your priorities, our execution.
Reporting: Monthly hours and progress, re-forecast when assumptions change.

Managed operations

Best for: Running what is already in production, including systems we did not build.
Scope: Systems, coverage window, service levels, exclusions and escalation rules.
Commercial: Quarterly or annual service fee, scoped to systems and coverage.
Ownership: NAFSoft operates; you retain business ownership and approval authority.
Reporting: Monthly service report: incidents, problems, changes, availability, actions.

Method

Discover → Specify → Build → Assure → Launch → Operate

Every model runs through the same six stages. What changes between models is who directs the work and how it is priced, never whether the work is specified and evidenced.

Quality gates we do not skip

Written scope and acceptance criteria before implementation

Peer code review; nothing merges without it

Automated tests plus a defined regression set per release

Security review against the OWASP Top 10 before production

Performance evidence at expected volume, not sample size

Cutover plan, rehearsal and a rollback that has been tested

Hypercare window after go-live with named responders

Handover pack: architecture, operations, runbooks, known risks

Describe the project and we will propose the right shape.

Include the deadline, the systems involved and the approval path. We will tell you which model fits, and where the delivery risk actually sits.

Talk to our experts