Unlocking the Secrets of Writing My Family History with ChatGPT's Surprising Questions

Perhaps your family history feels unexciting, even if you’ve gathered a mountain of research about your ancestors. Guess what? 

It might not be your sources.

Instead, you might not be asking the right question to improve it. 

Sometimes, an outsider—even a digital one—helps you see what you’re missing.

That happened to me, and I want to tell you about it.

My struggle to write about my Civil War Ancestor

I’ve spent years helping people write their family histories—on this blog, in workshops, and in my YouTube channel, Write Your Family History.

Even with my own helpful tips and tricks, I stalled while writing about my Civil War ancestor, William James Townsend. 

I used the Recipe for Writing Family History methods I teach to draft William’s story. I transformed records into paragraphs and expanded them with historical context. 

Yet, I lacked enthusiasm for formatting the project and preparing it for publication. What could be wrong? 

Watch the Video

A Family History Writing Conversation With AI

Now, as I’m sure you're aware, AI is all the buzz in genealogy circles. It seems RootsTech, YouTube, and Facebook formats are just flooded with tips and tricks.

I’m not opposed to using AI to help outline and overcome writer’s block, but I didn’t know what my sticking point was, so how could a language model help me?

I told ChatGPT about my Civil War Ancestor’s writing project and the overall outline. I explained the types of content I had, including documents, maps, photographs, and newspaper articles.

I didn’t give it a draft; I gave it my concept.

Following Stephen Little’s advice from RootsTech 2026, I spoke to ChatGPT in plain, conversational language: “I’m stuck trying to move forward. What questions should I consider to clarify my vision?”

Do you see?

I didn’t want ChatGPT, or any AI model, to write anything, but to help me figure out what I was trying to build.

The Question That Shifted Everything

After a few back-and-forth exchanges, I had a writing-changing moment.

ChatGPT asked: “Do you want each chapter to begin with scene-setting or fact-first narration?”

This question isn’t earth-shattering. In fact, it sounds rather simple.

Do I want the text to:

• Set the scene: establish time, place, sensory details, and emotional elements

• Hit the facts first: recount events in a chronological orientation that focuses on what’s happening, not where it’s happening.

I hadn’t deeply considered this at the chapter level for William’s project.

As you may know, I love the Zoom In technique to incorporate historical context and anchor a reader in a time and place.

However, I stopped and considered, is this the best approach for each chapter?

After some consideration, I realized:

• For William’s early life and post-war years, I do want to open with scene-setting. I want to bring the reader into his world.

• For his 90-day military service, I want a faster, factual pace. That section needs a different rhythm, so I’ll use fact-first openings there.

Once I recognized that William’s story demanded a varied approach, I found my flow again. I hadn’t initially intended to change the structure with each chapter, but the story itself required it. This realization released the blockage.

Follow-Up Questions After Finding the Sticking Point

What’s interesting is that whenever you find your sticking point, a flood of additional questions follows so that you can really define your vision.

That happened to me, and this time, no AI model was helping. My brain was firing on all cylinders, and I wrote a list of additional questions, such as:

• Do I want the same tone across the whole book?

• How long should chapters be?

• Will I include sidebars or stick to one narrative thread?

• How will I use images, if any?

So I sat down and started answering those questions, and the result was practically a vision board for William’s story.

Could AI Help Your Family History Writing Project Get Unstuck?

AI is a great tool when used correctly. I don’t believe AI should ever replace us as family historians. Many companies and genealogists claim their quick AI storytelling tools will rapidly write your family history. 

In reality, they rapidly write generic slop if you let them.

But if you use AI as an assistant, you can get unstuck and move your stories from your "someday" list to the finished pile. 

Always remember: we write so our families can connect to their ancestors, not so they roll their eyes and think, “Grandma’s telling boring stories again.”

In short, I wasn’t stuck on William’s story because I didn’t have enough research.

I was stuck because I hadn’t made a decision that would shape his story.

And what’s interesting, this sticking point won't be the same for the future stories I write, or the one you’re writing. It will be different for every story and for every writer. But AI can help get us unstuck if we use it wisely.

Now, in the next blog post, The 3 Layers of Decisions That Turn Genealogy Into a Story, I’ll share a few decisions you need to make after you’ve written your first draft so you can move to your final published manuscript quickly. 

In the meantime, can you let me know if you’ve ever gotten unstuck in your writing and what helps you?

Or, if you are stuck, what is your current sticking point?