AI Without Clutter: A Salvation Army Leader's Guide to Working Smarter
Introduction: Our AI-First Transformation
At G-Lab, we are embarking on a digital transformation—from a digital-first to an AI-first mindset. This shift means more than just adopting AI tools; it's about deeply integrating AI into how we work. Rather than seeing AI as a one-size-fits-all solution, we started taking literal notes on every work scenario where AI could help.
However, we quickly realized a challenge: how do we communicate to our AI and data teams which tasks to automate when we can't even map out our own workflows step by step?
Writing a blog post, reaching out to potential volunteers, planning a fundraising campaign—these processes don't follow a rigid checklist. They change every time. How could we recognize consistent patterns in what felt like unpredictable workflows?
That's where workflow documentation became the foundation of our AI-first strategy. In this post, we'll walk you through how we uncovered structured workflows in the middle of flexible, dynamic work—and how you can do the same to make AI truly work for you.
Step 1: Understanding What an AI Workflow Is
AI Understands Workflows Differently Than We Do
A workflow is a structured sequence of tasks that guide how work gets done, from start to finish. For Salvation Army leaders, workflows include everything from volunteer coordination to donor engagement and event planning.
AI, however, sees a workflow as a series of repeatable patterns and structured steps—it works best when tasks follow clear sequences. But real-world workflows are rarely that predictable.
For example, consider bellringer volunteer coordination. A leader might start by identifying volunteer needs, contacting available volunteers, and assigning roles. But in reality:
Schedules change last minute.
Urgent community needs arise.
Bellringers cancel, and new ones step in.
If AI is trained on a rigid workflow, it might assign volunteers based on past availability, not real-time constraints—leading to frustrating inefficiencies.
You need a system that helps you match the right volunteers with the right tasks efficiently without spending hours tracking them down.
To make AI truly useful, you must train it with structured workflows while still allowing for human flexibility. This means defining:
What are the core steps in this process?
Where do exceptions and last-minute changes usually happen?
How can AI assist without replacing human oversight?
When structured correctly, AI can assist by:
Sending reminders to volunteers.
Suggesting backup options for cancellations.
Analyzing participation trends to optimize future outreach.
Step 2: Documenting Your Workflow Before Using AI
The Challenge: AI Feels Generic Without Structure
Many AI tools create the illusion of progress without delivering real efficiency. That's because AI can only enhance what is already structured—it cannot create clarity where none exists.
The key to making AI work isn't picking the right tool—it's understanding how your work actually happens.
At G-Lab, we solved this by documenting our real workflows as they naturally happened instead of trying to force rigid step-by-step instructions upfront.
G-Lab Case Study: Using Screen Recording to Discover Hidden Workflows
Since workflows can be difficult to map manually, we used screen recording tools to track how we completed recurring tasks.
Here's what we did:
Start recording before beginning a task (e.g., scheduling volunteers, writing a report, planning an event).
Work naturally—as if there was no documentation process.
Review the recording later to analyze what steps we actually followed.
Summarize key takeaways in a document for reference.
Upload all documented steps to ChatGPT to analyze and refine workflow structures.
A real-world example: When I was working on creating new fundraising strategies, I used to think my process was completely different each time. However, after recording my screen during several strategy development sessions, I discovered that I unconsciously followed similar patterns. I would always start by reviewing past campaign data, then research donor demographics, sketch a timeline, and draft key messaging points—regardless of the specific campaign. This revelation helped me create a structured framework that still allowed for creativity but eliminated the "staring at a blank page" problem I often faced.
Try This Prompt in ChatGPT:
"I have documented my workflow for [specific task]. Here are multiple step-by-step records of how I completed this task over time. Can you analyze these and suggest the most effective workflow structure? Please optimize the process and recommend areas where AI could help streamline repetitive tasks."
After a few weeks of doing this, we uncovered real, structured workflows—not just vague ideas of how we thought we worked. This allowed us to refine processes and implement AI where it made the most impact.
Step 3: Implementing AI in a Targeted Way (Small Wins First)
The Mistake: Trying to Use AI Everywhere at Once
AI isn't meant to replace everything overnight. Many leaders try to automate too much too fast, leading to confusion and frustration.
The Solution: Start with Small, High-Impact Wins
AI is most effective when introduced gradually, improving one part of your workflow at a time.
At G-Lab, we started by using AI for simple, repetitive tasks first, such as:
Generating first-draft volunteer outreach emails.
Organizing data from donor reports.
Scheduling content for social media.
Our experience: One of our team members spent hours each week creating content calendars—a task that requires creativity and involves many repetitive elements. Instead of trying to automate the entire process, we focused on just the research phase. By creating a specialized custom GPT focused explicitly on content research, we cut the time spent on background research by 40% while still maintaining full creative control over the actual content creation.
Once we had small wins, we scaled AI's role strategically, using it to analyze workflow trends and suggest improvements.
Step 4: Maintaining & Refining with a Weekly "Spring Cleaning" Day
The Mistake: AI Workflows Get Cluttered Over Time
Without regular maintenance, AI knowledge bases become outdated and unstructured.
The Solution: Establish a Weekly AI Maintenance Day
At G-Lab, we are starting to implement the culture of Friday Maintenance Days, where we:
Review our most-used AI chats.
Extract key takeaways and update AI knowledge bases.
Clean up disorganized documentation.
Refine AI workflows based on what's working.
Why this works: Just like physical spaces, digital workspaces need regular tidying. During one of our Friday maintenance sessions, we discovered that several team members had created nearly identical custom GPTs for similar tasks. By consolidating these into a single, more robust assistant with a comprehensive knowledge base, we eliminated redundancy and improved consistency across our work.
This spring cleaning ritual is becoming one of our most valuable practices. It ensures our AI tools continue to evolve with our changing needs rather than becoming digital clutter.
Understanding Your Productivity Patterns
As we've implemented these workflows, we've discovered something even more valuable: insights into our own productivity patterns.
By documenting our processes, we've learned to distinguish between our "creative mode" and "execution mode" tasks. We've discovered when we work best on certain types of projects and how to structure our days to maximize both creativity and efficiency.
Personal insight: Through consistent documentation of my work patterns, I discovered that I experience significant frustration when tackling highly strategic tasks without proper structure. Now, I always create a visual roadmap through whiteboarding before diving into abstract strategic work, which has dramatically improved both my productivity and satisfaction.
What made this discovery so valuable wasn't just identifying a personal preference—it revealed a fundamental workflow obstacle. By carefully tracking and analyzing our stressors (not just our moods), we identified specific friction points in our processes. For instance, we realized that we had been structuring tasks inefficiently by combining creative and execution phases in the same day.
Our documentation showed that switching between these different mental modes was causing significant productivity loss. In response, we restructured our workflow to separate creative days from execution days, allowing team members to remain in a consistent mental state rather than constantly shifting gears.
Even the way we named and framed tasks needed adjustment. Instead of broad assignments like "complete content calendar," we began using more specific, phase-appropriate language like "research content themes" (creative phase) or "schedule approved content" (execution phase).
These discoveries weren't just interesting—they were transformative. By understanding our natural work patterns, we can now approach our AI and data teams with precisely defined automation needs rather than vague requests. We're in the early stages of this journey, still learning about ourselves and our processes, but each documentation cycle brings us closer to truly optimized workflows.
Conclusion: AI as a Long-Term Ally, Not a Quick Fix
Making AI work isn't about picking the right tool—it's about cleaning up your workflows first.
AI doesn't create efficiency—it amplifies it.
Structured processes lead to better AI results.
Spring-cleaning your workflow makes AI a true ally, not a gimmick.
By shifting to an AI-first mindset, we're not just adopting AI tools—we're transforming how we work, document, and optimize processes for lasting impact.
For Salvation Army leaders, this approach aligns perfectly with your mission of stewarding resources wisely while maximizing community impact. Start small, focus on understanding your processes, and watch as AI becomes a valuable partner in advancing your important work.
Struggling with a specific workflow or AI challenge?
Send us your question, and we'll feature answers to selected questions in our upcoming newsletter.
Ready to Start Spring Cleaning Your Workflows?
Here is our free "AI Workflow Prompts for Salvation Army Leaders" guide to jumpstart your process documentation journey. These role-specific prompts are designed to help you uncover the hidden structure in your daily work.
At G-Lab Group, we're committed to supporting Salvation Army leaders as you navigate digital transformation to advance your mission. Our expertise lies in helping local units tell their stories, and in an AI-first world, every story starts with well-documented and optimized workflows that make technology work for you, not against you.
Do you have a specific workflow challenge? Email us at eli.silva@g-labgroup.com for personalized suggestions. We're partners in your mission, dedicated to equipping you with the tools and knowledge you need to make a greater impact.
AI Workflow Prompts for Salvation Army Leaders
For Development & Resource Directors
Donor Engagement Workflow Analysis
Copy
I need to document my donor acknowledgment process. Currently, I handle thank-you messages, impact updates, and recognition for different donation levels. Analyze these notes from my workflow [paste your workflow notes] and help me identify: 1. Core steps that should remain consistent 2. Areas where personalization is most valuable 3. Which parts could be enhanced with AI while maintaining authentic relationships 4. How to structure this process better for both major donors and regular contributors
Donor Engagement Workflow Analysis
Copy
I'm planning our Red Kettle campaign workflow. Help me document the steps from planning to execution using these observations [paste observations]. Identify which planning stages could benefit from AI assistance and which require human creativity. How can I structure this workflow to allow for seasonal variations while maintaining consistency?
For Community Relations & Communications Managers
Content Calendar Optimization
Copy
I manage our local unit's communications across multiple channels. Review my current content creation process [paste process notes] and help me separate creative tasks from execution tasks. Suggest a workflow structure that would allow me to batch similar activities and identify where AI could help with research or draft content while preserving our authentic voice.
Crisis Communication Preparation
Copy
I handle disaster response communications for our local unit. Based on these past response examples [paste examples], help me create a structured workflow for crisis communications that incorporates both the necessary rapid response elements and thoughtful messaging review. Where could AI tools assist without compromising quality or accuracy?
For Operations Leaders
Volunteer Coordination System
Copy
I coordinate volunteers for our local programs. Here's how I currently manage recruitment, scheduling, and follow-up [paste current process]. Help me identify patterns in this workflow, suggest improvements to the structure, and recommend specific areas where AI could help with initial outreach or scheduling while maintaining personal connections with our volunteers.
Program Efficiency Analysis
Copy
I oversee our local unit's seasonal programs (food assistance, shelter services, etc.). Here's documentation of how we currently transition between programs throughout the year [paste documentation]. Help me identify inefficiencies in these transitions and create a more structured workflow that maintains quality of service while reducing administrative burden.
Using These Prompts Effectively
Document First: Before using any of these prompts, spend 1-2 weeks recording your actual workflow using screen recording or detailed notes.
Be Specific: Replace the [paste your workflow notes] sections with your actual observations, not just how you think the process works.
Start Small: Choose just one workflow to optimize first before moving on to others.
Review & Refine: Share the AI's suggestions with your team and adjust based on their feedback before implementation.
These prompts will help you begin structuring your workflows, making them clearer for both your team and any AI tools you implement in the future.