Cressida Takes Over — How to Research Like a Detective (Without Boring Your Reader to Tears)

Character Takeover Series | Entry 01 Cressida Vale has opinions. She has been given the blog. This was either a brilliant idea or a catastrophic one — possibly both.

---

Okay so first of all, hi. I'm Cressida. You can call me Cressi if we're friends, which we are now, congratulations, welcome to the inner circle.

I was told I could talk about whatever I wanted as long as it was useful to writers. So naturally I'm going to talk about the thing that makes me absolutely feral when I read fiction: research that was clearly done, clearly extensive, and then dumped directly onto the page like someone knocked over a filing cabinet in chapter three.

You know the scene. You're reading a mystery—or a thriller, or a historical romance, or a fantasy with a magic system—and the protagonist is just vibing, following a lead, being interesting, and then suddenly there are four paragraphs about the history of Victorian coroner practices and you, the reader, are somewhere between impressed and unconscious.

I feel this personally. I am a person who researches. Obsessively. Joyfully. Miles says I have a problem. Miles is not wrong, but that's beside the point.

The point is: research is the skeleton. The reader should never see the skeleton.

Let me break this down.

First: Why You're Info-Dumping (It's Not Your Fault)

Here's the thing about research that nobody warns you about. You do all this work—interviews, archives, three hours down a Wikipedia rabbit hole that started with "forensic accounting" and ended with "the surprising history of competitive cheese rolling in Gloucestershire"—and you love it. You love what you found. It's fascinating. You feel like a genius for knowing it.

And then you want the reader to feel like a genius too.

So you put it all in.

I understand this impulse completely. When I was working the crime beat, I'd do a week of research on something like small-town fraud schemes — the psychology, the financial patterns, the particular way people rationalize taking from people they know — and then I'd have to fit it into eight hundred words for a paper that mostly wanted the who and the when.

The discipline of what to leave out is a skill. It doesn't come naturally. It has to be built.

But here's what I've figured out, from years of having to condense enormous amounts of information into the thing that actually matters: your research is not the story. Your research is what makes the story feel true.

Those are completely different jobs, and confusing them is how filing cabinets get knocked over.

The Investigator's Rule: Only Show What the Scene Would Actually Surface

When I'm working a case—real or imagined—I know a lot more than I use at any given moment. I know the history of the property I'm standing in front of. I know the financial records of the person I'm about to interview. I know three different theories about what happened and which one I think is most likely.

But when I walk through that door, I'm not reciting all of that. I'm using it. It's in the way I notice which chair the homeowner sits in without being invited. It's in the specific question I ask that sounds casual but isn't. It's in the reason I already know the answer to the question before I ask it.

The research lives underneath the scene. It shapes what the character notices, what they ask, what they find significant. The reader feels the expertise without being handed the filing cabinet.

Here's a practical test: if you can remove an information paragraph from your scene and the scene still makes sense — if nothing is lost except the information itself — that paragraph is doing the wrong job. Information for its own sake is trivia. Information that changes what a character does, believes, or chooses next is story.

The "Cressida Method" (I'm calling it that, you can't stop me)

When I'm dropping research into a scene, I run it through three questions. You can steal these. I'm generous like that.

1. Whose knowledge is this?

Information should always belong to someone. It lives in a character's expertise, their background, their obsession, their fear. A forensic accountant notices the pattern in the numbers because that's how she's trained to see the world. A botanist notices the specific plant by the victim's door because her brain catalogues those things automatically. The information arrives through a lens, and that lens is characterization.

If your research feels generic—floating above the scene without belonging to anyone's specific perception—it's not integrated yet. Ask: who knows this? Why do they know it? What does it cost them or reveal about them to know it?

2. Does this change something?

Research that creates story momentum is earned. Research that pauses story momentum is a detour. There's a difference between:

"She knew that arsenic poisoning had been commonly used in Victorian England because it was undetectable in most post-mortems at the time, and the medical establishment was only beginning to develop reliable testing methods—"

And:

"Arsenic. Undetectable in 1870, practically speaking. Which meant someone had known exactly how long they had before anyone started asking the right questions."

Same information. Second version: belongs to the character, moves forward, creates implication. First version: is Wikipedia dressed up in novel clothes.

3. Can it live in action rather than explanation?

Show me the detective running her finger along a dusty shelf and frowning — because she knows what that specific pattern of dust disturbance means — rather than explaining what she knows about evidence preservation. Let the reader watch the expertise operate and then catch up to why it mattered.

Action-first, explanation only if necessary, and even then: brief. Surgical. One sentence that unlocks the door rather than three paragraphs that describe the door in loving detail.

A Note on Dialogue as Research Delivery

Conversation is the most forgiving container for information because it has friction built in. Characters can interrupt each other, misunderstand each other, disagree. The information has to fight to exist in dialogue, which automatically keeps it from sprawling.

"How do you know that?" is one of the most useful sentences in research-heavy fiction. It lets another character ask for the filing cabinet so your protagonist can open it briefly, deliberately, and then put it away.

But even in dialogue: don't let your character become a lecture. Nobody — not even a historian, not even a pathologist, not even a detective who is very enthusiastic about the particularities of arsenic poisoning — explains everything they know at once. They answer the specific question. They follow the conversational thread. They circle back to the thing that actually matters.

What Well-Integrated Research Actually Does

When you get this right — when the research lives under the scene instead of on top of it — something magic happens.

The reader doesn't feel informed. They feel immersed.

There's a difference. Being informed is intellectual. Immersion is physical. It's when you smell the archive, feel the weight of a specific era, understand intuitively how this world operates because the character moving through it knows it in their bones and carries that knowledge in their choices and their noticing and the things they don't have to say.

That's the goal. Not a reader who knows more after your scene. A reader who was there.

Research is how you build a world the reader can inhabit. The reader never needs to see the blueprints.

Trust that you know more than you're showing. Trust that the iceberg is working. Trust that the reader will feel the depth even when they can't see the bottom.

Now if you'll excuse me, I have a case file and approximately four unanswered questions that are going to bother me until I answer them.

(Miles says I should add a conclusion. This is the conclusion. You're welcome, Miles.)

— Cressida Vale

---

Want more mysteries in your inbox?

Join The Blackvale Dispatch for new cases, behind-the-scenes clues, exclusive bonus content, and the occasional cryptic message that may or may not be relevant to future investigations. Cressida approves. Miles is suspicious but willing.

👉 Join the Blackvale Dispatch

No spam. No nonsense. Just mystery.