How we work
Clear decisions. Working software. Accountable delivery.
Our delivery approach connects business requirements, engineering decisions and operational ownership. You review working software throughout the project, with clear milestones and responsibilities.
Scope around a complete user journey
We begin with the work people need to complete and the systems, data and constraints around it. Custom development, configuration and integration are assessed against those requirements before we recommend an architecture.
The first-release scope identifies users, priority journeys, exclusions, dependencies and acceptance criteria. Where a legacy system or integration creates uncertainty, a focused technical investigation informs the delivery plan.
Keep delivery decisions visible
We work in focused iterations, with regular reviews of working software. These reviews give the people using and funding the system a basis for assessing progress and deciding priorities.
The engagement defines who owns those priorities, how scope changes are approved and where decisions are recorded. Each review establishes what has been accepted, what needs revision and the next milestone.
Match engineering controls to the application
Permissions, testing, integrations, migration and release preparation are part of the delivery scope. We agree the application’s performance and recovery requirements and the checks needed to assess them.
We build to WCAG 2.1 Level AA as the default target, for public pages and for the screens your staff use. Testing covers keyboard-only completion of the agreed journeys, visible focus, form errors that name the field and the fix, colour contrast, and a screen-reader pass on the journeys that matter most. Anything we have not resolved is listed at handover rather than left for you to find. Where a funder, a government contract or provincial legislation sets a different bar, we scope to that instead.
Depending on the project, release evidence can include automated test results, reviewed user journeys, migration reconciliation and a tested restore procedure. Independent security assessments and specialist testing are specified separately where required.
Give AI its own acceptance criteria
AI features need representative examples, a definition of acceptable output and a response to errors. We consider the surrounding application alongside the model: data access, action limits, review steps, usage costs and fallback paths.
Our AI production guide explains how to evaluate a pilot before extending its use. The AI evaluation worksheet provides a record for the task, test cases and release decision.
Engineering decisions in practice
Healthy Aging CORE brings regional publishing and membership permissions into search and collaboration. CareOps connects project activity to AI-assisted report drafts and staff review. Networkbase combines member context with outreach proposals and campaign drafting.
Each project illustrates how the architecture, interfaces and access rules support a particular workflow.
Plan the handover before launch
You own the custom code we write for you. On payment in full, all right, title and interest in the deliverables — the source code, configuration, documentation and scripts created for your project — belong to your organization, and we hand over the source into a repository you control. Your data is yours throughout, not only at the end.
Where a project reuses a component we already owned, we keep that component and grant you a perpetual, royalty-free licence to use, modify and maintain it as part of your system. Third-party libraries stay under their own terms. We tell you which parts those are during the project, so you know what you own outright and what you hold under licence before you commit — not when you come to leave.
Infrastructure accounts, domains, data and third-party services are named in the agreement rather than assumed, along with the documentation and operational walkthrough required for handover.
Support arrangements define coverage, availability, escalation and the handling of ongoing changes. Project-specific support or the Software Care Plan can be discussed before release.
Our discovery guide and handover checklist outline the decisions and materials involved. Use the project brief to prepare your requirements for an initial conversation.
Keep data in Canada
For clients in Canada, we host on infrastructure in Canadian regions, so the records the system holds stay under Canadian jurisdiction. Organizations handling member, client, health or donor information are usually asked where that data lives — by a funder, an insurer, a privacy officer or their own board — before a system is approved.
Where a client operates in another country, we host in the jurisdiction their own obligations require. Either way, if a project needs a third-party service hosted somewhere else, we identify it during scoping rather than after launch, and agree what data reaches it and why.
Work with a Vancouver-based team
RapidIteration is based in Vancouver, British Columbia, and works with organizations across Canada. Communication, review cadence and decision owners are agreed with each client, so the delivery process fits the people responsible for the project.
Discuss your next project
What does your
organization need next?
Tell us about your priorities, current systems and the change you want to make.
We’ll help define the scope and approach.