Skip to content

Demo content Everything on this site is sample data shipped with the template. Configure it or follow the launch guide (opens in a new tab).

In production · Metro North Health District

Vendor product

Grant narrative copilot

Result in one line: Halved first-draft time on routine narrative sections, from about eight hours to four

A vendor-hosted assistant grounded in our own program library, used by staff to draft first-pass narratives for federal funding applications.

Published April 14, 2026

Documentation or write-up
Skills needed to set it up
Power user
Readiness
Needs a paid license Needs customization Human review built in
Data it touches
Internal, non-public data
Who sees the output
Internal staff
License
Not open source — description only
Cost to stand up
Under $25k
Portable to other platforms
Partially — with rework
Use case category
Communications, media & writing
Reviews it went through
Not yet reviewed
What do these mean?
Power user
Someone comfortable with spreadsheets, forms and low-code tools.
Needs a paid license
Depends on a paid product, API or subscription.
Needs customization
Your team will need to adapt it before use.
Human review built in
A person checks the AI's output before it is used.
Internal, non-public data
Non-public operational data without personal identifiers.
Internal staff
Used only inside the organization.
Not open source — description only
The write-up is what is shared, not the artifact itself.
Partially — with rework
Some pieces are vendor-specific and would need swapping.
Communications, media & writing
Drafting, translating, summarising and publishing.
Not yet reviewed
Shared honestly — this has not been through a formal review.

Problem

We apply for between 15 and 25 federal and state funding opportunities a year. Roughly 60% of every narrative is material we have written before: organizational history, population profile, governance structure, evaluation approach. Staff were rewriting it each time, usually under deadline, and the versions drifted apart until nobody could say which description of our own health district was current.

What we built

There is no code here. This is a configuration and process story, which we think is most of what other organizations actually need.

We assembled a SharePoint library of about 120 approved items: program descriptions, evaluation summaries, boilerplate paragraphs, and the population data tables we are allowed to cite. Each item has an owner and a review date. We then pointed the vendor’s assistant at that library and only that library. Staff paste the section headings from a funding notice and get a first draft written in our own language.

How it works

The assistant is licensed per seat and runs on the vendor’s infrastructure. Nothing outside the approved library is in scope, and no application draft is stored anywhere other than the working folder for that application. Every draft goes through a fact-check step before it is submitted — that checklist is linked above and it is the single most important part of this entry.

Results

First-draft time on routine sections fell from roughly eight hours to four. More usefully, the boilerplate stopped drifting: because the library is the only source, updating a program description once updates it everywhere the next time someone drafts.

Lessons learned

The assistant is confident about numbers it has never seen. Early drafts contained plausible-looking population figures that appeared nowhere in the library. Every statistic is now verified against the source document before submission, and we added a rule that the assistant may not produce a number the drafter has not already supplied.

Novel program designs still start from a blank page. The tool helps most with the 60% that is standard and barely at all with the part that wins the award.

How to reuse

Build the library before you buy the license. A well-curated set of 100 approved documents with named owners is what makes this work; the assistant is the cheap part. Budget time for a review cycle on every item you put in, because whatever is in the library becomes what your organization says about itself.

About

Area of work
  • Leadership & administration
  • Policy & planning
  • Finance, procurement & contracts
Review status
Reviewed & approved

How it's built

How AI is involved
AI is part of the solution
Types of AI
  • Document Q&A (RAG)
  • Generative text (LLM)
AI tools & models
  • Microsoft 365 Copilot
  • SharePoint
Where it runs
  • Enterprise AI workspace
  • Vendor / SaaS hosted
Vendor or partner
Microsoft

Sharing & licensing

Access terms
Built on Microsoft Copilot inside our tenant; the prompt library and the reviewer checklist are available to peer health departments on request.
Portability notes
The prompts and the program description library are text; the retrieval over past applications relies on SharePoint indexing and would need re-wiring on another platform.

What it took

Cost to keep running
Under $10k/yr
How it was bought
  • Sole source
Who it affects
Nothing resident-facing: it drafts our own narrative sections and a grant writer rewrites them. The honest risk is internal — a copilot trained on our past successful applications will keep proposing the programs we have always funded. We have not checked for that and would not know how yet.

Data & access

No PII/PHI in the shared material
Yes
Data sources
  • Program description library
  • Prior grant applications (redacted)
  • Population data tables
Data-governance caveats
Past applications and program descriptions are internal documents; nothing containing client or patient detail is loaded into the library.