Scaling Up — First Drafts and Prospecting

Scaling Up

Your team has been using AI for editing and research. They’re catching errors, sharing learnings, and building confidence. Now you’re ready to take on tasks that score higher on one or more of the risk dimensions.

More Complex Analysis

These tasks involve lower context-to-output ratios and larger outputs — AI is generating analysis and judgment, not just processing what you gave it.

Funder matching. Ask AI to evaluate potential funders against your organization’s profile and priorities. AI can score fit across multiple dimensions — mission alignment, geographic focus, funding range, past giving patterns — faster than manual research.

Pipeline analysis. Use AI to analyze your grant pipeline — win rates, common rejection reasons, patterns in successful applications, upcoming deadline clusters.

Opportunity briefs. Have AI generate brief summaries of potential opportunities for leadership review — one-page overviews that help your ED or board decide where to invest time.

What to watch: AI is making analytical judgments that influence real decisions. Your team should evaluate AI’s analysis against their domain knowledge. “AI says this funder is a strong match” should be followed by “and here’s why I agree (or don’t).”

First-Draft Writing

This is the big one. AI producing first drafts of proposal sections is where the largest time savings live — and where the risks are highest.

You’re ready when your team:

• Has logged multiple cycles of AI-assisted research with successful review
• Can articulate the specific risks of AI-generated content (fabrication, bias, voice)
• Has a defined review workflow (who reviews, what they check, in what order)
• Trusts the review process enough that errors get caught

First-draft tasks:

Needs statement drafts. Provide AI with your outcomes data, community data, and the funder’s priorities. Ask for a first draft. Then: verify key claims, check the framing for bias, and adjust the voice.

Methods and evaluation sections. These are more formulaic and often benefit from AI’s ability to structure logical frameworks quickly.

Budget narratives. AI can draft justifications for line items based on your actual budget figures and program design.

Letter of inquiry drafts. Shorter pieces where AI can generate a solid starting point in minutes.

Helpful practices for first-draft work:

• The Spot-Check Technique (Track C, Lesson C4-03) is especially useful here. Walking through key claims against their sources catches the most common issues.
• Voice review matters. If you’ve set up style guides or organizational rules (as we covered in the bias lesson), AI should already be close — but check that it sounds like your team, not like a generic model.
• Remember the input-to-output principle: the more source material you give AI, the better the drafts. Upload outcomes reports, past proposals, and funder data before asking for a draft.

• Budget review time realistically. AI saves significant writing time, but you’ll reinvest some of that into review and refinement.

High-Stakes Content

Compliance-sensitive content — federal reporting, financial certifications, and legally binding language — carries the highest stakes. AI can be genuinely helpful here, and may even make the process safer by catching inconsistencies a human might miss. But close human sign-off is essential.

If you use AI for compliance content:

• Have someone with specific compliance expertise review the output
• Verify AI’s interpretation of requirements against the actual regulations
• Use AI as a second pair of eyes alongside human review, not as a replacement for it
• Consider using AI to check human-written compliance content as well — it works both ways

The Expansion Pattern

The pattern for scaling up is always the same:

1. Try a new task type with one team member or one project
2. Review the results carefully — what worked, what didn’t, what nearly slipped through
3. Share the learnings with the team
4. Adjust your process based on what you learned
5. Expand to more team members or more projects

Don’t rush it. The organizations that expand AI use steadily and learn from each step build more durable capabilities than those that try to adopt everything at once.

### Next Lesson

You can roll out AI and expand its use. But sustaining AI adoption requires leadership — building the culture, managing the skeptics, and measuring what matters. Module 5 covers leading AI-native teams.