The problem
Policy documents, review dates and staff acknowledgements can be scattered across Word files, spreadsheets and email, making the current approved version difficult to follow.
OneCare Policy
In developmentA policy-management platform I’m building as part of OneCare Digital’s work for independent homecare providers in England, bringing documents, reviews, approvals and staff acknowledgements into one workflow.
A working preview is available and the launch list is open. The platform is not yet generally available. AI-assisted checks and drafting are planned, with human review.
- My role
- Founder, designer and developer
- Platform
- Web application
- Built with
- Next.js, TypeScript, Supabase, Tiptap
My approach
I’m keeping the document, version history and review decisions together, with a manager responsible for approval.
The result
A working product preview, an original draft template library and a live launch website. Development and preparation for early access are ongoing.
Decisions
In care operations, I saw how policy documents, review dates and staff acknowledgements could become disconnected. A folder of files doesn’t show which version has been approved or who has acknowledged it.
I’m building OneCare Policy through OneCare Digital for independent homecare providers in England. The review process connects a policy’s wording with the people who approve and use it.
Start with one care setting
I’ve focused the language, templates and workflow on independent homecare providers in England. That gives me a specific care setting to design for and test with providers.
Make the approved version clear
I’ve given drafts, reviews and approvals distinct states, and kept the document history beside the current wording so managers can follow a change.
Support different starting points
Providers can start with an original draft template, import a document they have permission to use or write from scratch. They still need to adapt and review the policy for their service.
Keep managers in control
I’m planning AI-assisted checks and drafting, with human review. The provider will remain responsible for reviewing and approving a policy.
The work
A workspace around the document
I’m building the application with Next.js and Supabase, using Tiptap for the editor. The working preview includes the policy library, document imports and review decisions.
An original draft library
I prepared 60 original draft policies and 60 matching forms. Providers need to adapt, review and approve each document for their own service.
Review the wording that changed
The development preview puts added and removed wording beside the policy’s owner, current version and next review date. The manager can inspect the changes before choosing whether to approve the draft.
More implementation detail
A clear introduction before launch
I designed and built the public website with product walkthroughs and a launch list. It explains the working preview and the features still in development.
Outcomes
Working product preview
A working preview of policy editing and review, available while I prepare the product for early access.
Original draft templates
60 policy drafts and 60 matching forms prepared as starting points for provider review and customisation.
Public launch website
A live introduction to the product with a launch list for interested homecare providers.
Constraints, lessons and next steps
Constraints
- The product is in development; customer adoption and operational outcomes have not yet been established.
- Templates need provider customisation and review before use. The platform does not guarantee regulatory compliance.
- AI-assisted checks and drafting are planned. Any future suggestions must remain reviewable and separate from a manager’s approval.
Reflections and next steps
- My next steps are to evaluate the workflow with providers and prepare the product for early access.
- AI-assisted checks and drafting remain planned work, with human review required before suggestions are approved or used.
- My care-sector experience gave me a starting point. I still need to test the workflow with providers using it.
- I need to make the distinction between a suggestion and an approved policy clear, particularly as I develop AI assistance.