How to Prepare Horror Investigation Clues That Haunt

Make them memorable before they become meaningful

The investigators hear the melody through the wall of a cheap motel.

Five notes.

Nothing beautiful. Nothing dramatic. It sounds like somebody trying to remember the beginning of a children’s song and never quite reaching the next line.

One of the players asks whether they recognise it.

No.

They keep searching the room.

Forty minutes later, nobody is talking about the melody anymore.

Good.

Three sessions later they are listening to an interview recorded in 1974. A frightened woman describes something she saw beneath the railway arches. Her voice breaks. Someone off microphone tells her to take her time.

Then she begins to hum.

Five notes.

One player stops the recording.

“Play that again.”

Now we have a horror clue.

Not because the Game Master successfully hid information.

Because something that had already entered the players’ memories has just changed its meaning.

That is the sort of clue that can haunt a table.

A clue does not need to look like a clue

Investigation advice often begins with information.

What happened?

Which evidence points towards it?

How many routes do the investigators have to the next revelation?

Those are necessary questions. A mystery still needs causal foundations. The world has to know what happened even when the players do not.

But horror can do something more with evidence.

A clue can arrive before the players understand that they are investigating it.

A melody.

A smell.

A particular sentence.

A company’s logo.

A photograph with one suspiciously empty space.

A lift that pauses briefly between floors.

A child drawing the same black shape in the corner of every picture.

The first encounter does not have to say:

THIS MATTERS.

In fact, it can be stronger when it does not.

> A horror clue does not have to announce itself as evidence. It has to earn a place in the player’s memory.

That changes preparation.

Instead of asking only, “What information does this clue provide?”, ask another question first:

Why might somebody remember it?

Not because it is screaming.

Because it is specific.

Slightly wrong.

Unfinished.

Recognisable enough to stick.

Incomplete enough to worry about later.

> A haunting clue is complete enough to remember and incomplete enough to worry about.

That is the earworm principle.

You do not necessarily understand it.

You simply cannot quite get rid of it.

Hide the clue in life, not behind obscurity

There is a danger here.

Experienced players notice when the Game Master suddenly describes one completely unnecessary object with loving precision.

There are three items in the room.

A desk.

A chair.

And an exquisitely painted porcelain rabbit with one blue eye and one green eye.

Nobody believes the rabbit is decorative.

The poor thing might as well be wearing a sign saying IMPORTANT PLOT OBJECT.

If clues are going to enter the players’ minds before revealing their importance, the surrounding world needs enough life to hide them naturally.

Not hide them behind impossible perception rolls.

Hide them in normality.

A radio plays while someone searches a drawer.

A nurse complains about a tune a patient will not stop humming.

A company logo appears on the side of a delivery van.

An old prayer is printed in the margin of a funeral programme.

A child draws while the adults talk.

All of these things can simply belong to the scene.

> Hide the clue in life, not behind obscurity.

This is where rich worldbuilding helps investigation.

When the world contains ordinary detail, the extraordinary does not need a spotlight.

It can sit quietly beside everything else until the players learn how to see it.

The onion: let the same clue acquire deeper layers

Call of Cthulhu has always understood something useful about horror investigation: truth often works like an onion.

The investigators do not simply move from ignorance to answer.

They peel.

One explanation reveals another beneath it.

Then another.

And the most disturbing revelation is often not the last fact by itself, but the realisation that what seemed understandable three layers ago was never what they thought it was.

Our five-note melody can work the same way.

At the outermost layer, it is atmosphere.

A cheap motel. Thin walls. A radio somewhere next door.

Five notes.

Later, it becomes recognition.

The same melody appears on a recording from 1974.

Now the players know that the first scene contained something they did not understand.

The next layer gives context.

The recording belongs to an investigation into disappearances near the same railway line connected to the present case.

Now it is a pattern.

Another layer makes it human.

Several people who disappeared had hummed the melody in the weeks beforehand. One surviving witness remembers her sister doing it over breakfast and becoming furious when asked where she had heard it.

Now the clue carries grief.

Another layer reaches history.

The melody appears in handwritten notes from the 1920s, but the researcher who recorded it insisted that it was not a folk song. The intervals were wrong.

Another layer makes the question worse.

A musician examines the recording and says there is something peculiar about the sequence. It is structured less like music than like a repeated signal.

The clue has not changed.

The understanding has.

> The clue stays the same. Understanding does not.

This is the important part of the onion.

Each layer should not merely add another fact.

It should alter the meaning of what came before.

The motel scene changes.

The 1974 interview changes.

The surviving witness changes.

A scene the players thought they understood becomes infected by new knowledge.

> The best horror clues can change the meaning of scenes that have already happened.

That is far more unsettling than simply discovering another locked drawer containing another diary.

Layers deepen the clue. Resonance spreads it.

The onion gives us vertical movement.

We go deeper into what the clue means.

But a living campaign also gives us horizontal movement.

The clue can travel.

The melody appears in a motel.

Then an archive.

Then a witness statement.

Then an old family recording.

Then somewhere else entirely.

That is resonance.

> Layers deepen the clue. Resonance spreads it.

Those are different functions.

Layers ask:

What is this?

Resonance asks:

Where does it travel, who encounters it, and what happens because of that contact?

The melody does not need to repeat identically.

In fact, it should not.

Repetition gives recognition.

Variation gives meaning.

> Repetition creates recognition. Variation creates meaning.

One witness hums it.

Another writes the notes down as numbers.

An old newspaper describes children singing “an unpleasant little tune” after a local disaster.

Someone has incorporated its rhythm into a prayer.

A software engineer notices a suspiciously similar pattern hidden in a corrupted audio file.

Now the clue does not feel like a marker placed by the Game Master.

It feels as though it belongs to the world.

Let different people understand different layers

A clue can have one existence and still mean different things to different people.

That matters enormously in horror.

A police officer hears an odd recording.

A musician hears an impossible interval.

A historian notices that the recording technology does not match the date on the archive box.

A local grandmother recognises something her mother used to forbid the children from singing.

A cultist recognises the first four notes and refuses to hear the fifth.

A child simply likes the tune.

Nobody needs to possess the complete answer.

In fact, it is usually better when nobody does.

> The clue has one existence, but different people can occupy different depths of understanding.

That makes NPCs part of the investigation rather than machines that dispense exposition.

A witness gives you what they noticed.

A scholar gives you what their training lets them recognise.

A frightened parent gives you what the event meant to their family.

And each may be wrong about part of it.

The priest may interpret the melody as a religious warning.

The doctor may believe it is neurological.

The historian may assume cultural transmission.

The cultist may possess the most useful information and the most distorted explanation.

The players have to live between those versions.

That is where investigation becomes interpretation.

Behaviour can be a clue

This also means clues need not be objects.

A person can carry evidence in the way they behave.

Imagine a caretaker who answers every question politely.

Names.

Dates.

Visitors.

Missing tenants.

Everything.

Except Tuesday.

Ask about Tuesday and he does not freeze dramatically or start sweating under a spotlight.

He simply answers the question beside it.

“What happened on Tuesday?”

“Well, Wednesday was when the police came.”

“When did you last see her?”

“Monday evening. She usually came home late.”

“What about Tuesday?”

“She was a very private woman.”

At first, perhaps it merely feels odd.

Three sessions later, the investigators realise every relevant disappearance took place on a Tuesday.

Now that old conversation comes back.

> Behaviour creates evidence, not exposition.

The same can happen with a place.

A hospital lift has no button for B3 but occasionally stops there.

Every old house in one village has a second chimney that nobody uses.

A church contains five alcoves although only four saints are represented.

The first time, these things are worldbuilding.

Later, they become evidence.

> A clue can begin as worldbuilding and become evidence only when the players learn how to read it.

That is one of the great pleasures of horror investigation.

The world was telling you something before you knew its language.

Research tells us what belongs

This is where good research becomes particularly useful.

Horror often works through deviation.

To make deviation convincing, we first need to know what belongs.

Architecture.

Funeral practices.

Religious customs.

Children’s songs.

Professional routines.

Family traditions.

Medical procedures.

Local folklore.

Administrative language.

Transport networks.

Old maps.

Company structures.

The details themselves are not horror.

They establish the normal layers.

Then we move one thing.

Only slightly.

A local burial custom uses coins.

Fine.

Research tells us why, when and how.

Then investigators notice one family uses a different coin.

Still plausible.

Then they discover several examples carry dates decades later than the burials.

Now something is wrong.

Later they learn the coins were never offerings to the dead.

They were seals.

> Research tells us what belongs. Horror begins when one thing almost belongs.

That “almost” is valuable.

An enormous horned demon standing in the kitchen is unmistakably supernatural.

A family custom that has been practised incorrectly for four generations is harder to dismiss.

The investigators have to keep looking.

Make it personal without turning every Bond into a hostage

Eventually, some clues should be allowed to leave the case.

They cross into the characters’ lives.

That is when resonance becomes personal.

Suppose one investigator’s husband works at a hospital.

Weeks after the motel, over breakfast, he says:

“We had the strangest patient yesterday. Little girl. Wouldn’t sleep. Kept humming the same five notes all bloody night.”

He carries on buttering toast.

He has no idea what he just said.

The player does.

Nobody has kidnapped him.

Nobody has threatened the family.

The Game Master has not smashed a monster through the kitchen window.

The clue has simply crossed a boundary that the player thought was safe.

> A clue becomes personal when it crosses from the mystery into something the character already cares about.

That can be far more effective than automatic tragedy.

The same principle can reach into biography.

During character creation, a player established that their father used to tell them a particular story.

Ten sessions later they find a different version of that story in an archive.

The ending is wrong.

Or perhaps their father’s version was wrong.

Why?

Was it a family variation?

Did he know something?

Who taught it to him?

Now the investigation no longer asks only:

What happened here?

It asks:

What does this have to do with me?

And sometimes:

What does this mean about me?

That is where horror becomes intimate.

Some clues cost nothing until you understand them

Horror advice often recommends attaching a price to knowledge.

That can work.

The forbidden book damages Sanity.

The recording attracts something.

The archive search exposes an identity.

The ritual requires the investigators to involve someone dangerous.

But if every clue bites the hand that touches it, players learn a simple lesson:

Stop investigating.

There is another kind of cost.

Understanding.

> Some horror clues cost nothing when you find them. The cost comes when you understand what you already saw.

The photograph was harmless when you found it.

Now you know why someone was cut out of it.

The nursery rhyme was irritating.

Now you know where it came from.

The neighbour seemed eccentric.

Now you understand what she meant when she told you that the house “doesn’t like new children”.

Nothing has physically attacked the investigator.

Knowledge has altered the world they thought they occupied.

That is very Call of Cthulhu.

The truth is not frightening merely because it contains monsters.

It is frightening because it changes the meaning of reality.

Give them the evidence. Let recognition do the haunting.

This only works when players are allowed to find things.

Do not hide the five-note melody behind a meaningless failed Listen roll.

If the characters are standing beside the radio and listening to it, let them hear it.

If they inspect the photograph, let them see what is in the photograph.

If they ask a plausible question, give them the base evidence.

Rolls can determine depth.

Precision.

Speed.

Risk.

Professional interpretation.

What else the character notices.

What their investigation disturbs.

But essential evidence should not disappear because a die rolled badly.

> Give them the evidence. Let recognition do the haunting.

The interesting challenge is not whether the players are permitted to possess information.

The interesting challenge is what they make of it.

World creates evidence.

Players create meaning.

And players will sometimes create the wrong meaning.

Good.

A false interpretation can still produce real consequences.

They may accuse the wrong person.

Trust the wrong witness.

Protect something they should fear.

Destroy something that was keeping another danger contained.

The mystery moves because people act on what they believe.

That is more alive than waiting for everybody to solve the Game Master’s intended riddle.

Do not turn every repeated detail into the answer

There is one danger with resonant clues.

Players learn quickly.

If everything that appears twice is secretly connected to the central horror, they will begin treating recurrence as a mechanical flag.

The second appearance of a red umbrella immediately produces:

“Right. Cult symbol.”

That becomes another kind of metagame.

A living world needs texture.

Patterns can be meaningful.

They can also be cultural.

Coincidental.

Misunderstood.

Deliberately manufactured.

Or created by the investigators themselves because human beings are very good at finding patterns once they start looking.

This is not permission to cheat.

The world still needs causal integrity.

But not every echo needs to lead to the same basement.

Some similarities should remain similarities.

> Similarity is a finding, not automatically an explanation.

That preserves uncertainty.

And uncertainty is where horror breathes.

The deepest layer should change the earlier ones

Eventually, the investigators may discover what the melody actually is.

Perhaps it is not music.

Perhaps human hearing translates something into tones because that is the nearest pattern the brain can tolerate.

Perhaps it is a signal.

Perhaps it is an address.

Perhaps it is the response of something enormous to a stimulus human beings do not know they are producing.

The specific answer matters less than what it does to everything before it.

The motel was no longer random.

The 1974 recording was not simply historical curiosity.

The missing people were not merely victims.

The father’s old story was not merely a family story.

Every layer rotates slightly.

That is what the final revelation should do.

Not:

“Here is the monster.”

But:

> “Then what the hell have we been looking at all this time?”

The strongest horror revelation expands backwards.

It contaminates memory.

And then let the clue return once more

The case may eventually end.

Perhaps the investigators destroy something.

Perhaps they understand enough to contain it.

Perhaps they run.

Perhaps they are completely wrong.

Whatever happens, resist the urge to explain everything.

Months later, one of them comes home.

Breakfast.

Coffee.

Their son is staring at a phone while spreading far too much jam over a piece of toast.

He is humming.

Five notes.

The investigator stops.

“Where did you hear that?”

The boy looks up.

“Hear what?”

Silence.

Then he goes back to breakfast.

That is enough.

Because by then the Game Master no longer needs to tell the player why those five notes matter.

The player brought that fear with them.

That is the goal.

Not to hide the clue.

Not to defeat the investigators.

Not to construct a puzzle clever enough to prove that the author was clever.

Give the players something they can notice.

Let it find a small place in their memory.

Bring it back differently.

Peel another layer.

Let it move through people and places.

Allow it eventually to touch something personal.

Then let new knowledge change what the players thought they already understood.

> Make it memorable before making it meaningful.

> The clue stays the same. Understanding does not.

> Layers deepen the clue. Resonance spreads it. Personal connection makes it hurt.

The strongest horror clue is not necessarily the one the players finally understand.

It is the one whose newest layer makes them afraid of everything they understood before.

And the best response at the table is rarely:

“We found the clue.”

It is:

“Wait. We have seen this before.”

That is when evidence stops being information.

That is when it starts haunting the table.

MYTHVEIL

We create worlds people want to live in.

Stories that remember. Worlds that change.

Read: https://mythveilchronicles.com/ Play: https://startplaying.games/gm/dunchan Watch: https://www.youtube.com/@MythveilChronicles Support: https://www.patreon.com/MythveilChronicles Collaborate: https://www.linkedin.com/in/francisco-jos%C3%A9-z-b05aab371/