If your dialogue sounds fake, it’s almost always because your characters agree on what the scene is about. They’ve read the same script you have. They know what the conversation is for, they know where it’s going, and they’re helping you get there.
Real people never do this. Two people in a room almost never agree on what a conversation is actually about, and that disagreement is what keeps people reading.
What does agreeing on the scene actually look like?
Here’s the pattern. Character A wants to talk about the missing money. Character B also wants to talk about the missing money. They sit down. They talk about the missing money. The scene ends.
Nothing is wrong with any single line. The words are fine. The problem is that both characters showed up to the same scene, with the same agenda, and the same understanding of what’s on the table.
We read that and it feels flat because we know instinctively that people don’t work this way. When your partner wants to talk about the holiday budget, you want to talk about why they left the back door open again. Two scenes are happening at once, and neither of you is willing to admit it.
That collision is what makes dialogue interesting. The polite version, where everyone stays on-topic, is boring.
Why do writers fall into this trap?
Because we know what the scene is for. We wrote the outline. We know this is the confrontation scene, or the reveal scene, or the reconciliation scene. And so we hand both characters the outline and let them cooperate.
The result reads like two actors doing a table read of a script neither of them wrote. Technically accurate. Emotionally dead.
The other reason is efficiency. If both characters agree on the topic, the scene moves fast. You get from the top of the page to the reveal in half the words. That feels good when you’re drafting. It feels terrible when you read it back.
Dialogue is not supposed to be efficient. It’s supposed to be human, and humans are wildly inefficient when they talk to each other.
How do you write characters who disagree about the scene itself?
Give each character a private goal for the conversation, and make sure those goals don’t match.
Before you write the scene, answer three questions for each character in it:
- What do they think this conversation is about?
- What do they actually want out of it?
- What are they trying to avoid talking about?
If both characters give you the same answers, you don’t have a scene yet. You have a summary. Change one of them.
One character might think they’re having a conversation about the missing money. The other might think they’re having a conversation about whether they can still trust each other. Both are valid. Both are urgent. And they will keep talking past each other in a way that feels electric on the page.
A quick example of the shift
Weak version: A asks about the money. B explains where it went. A asks a follow-up. B answers. The scene resolves.
Stronger version: A asks about the money. B answers a question A didn’t ask, because B is trying to steer toward the trust issue. A gets frustrated and pushes harder on the money. B goes quiet, then says something about last Tuesday that A doesn’t understand. Now we’re reading.
Same information. Same beats. Completely different scene, because the characters are no longer cooperating with the writer.
What about scenes where characters really do agree?
There are scenes where two characters want the same thing and are aligned on the topic. Reconciliation scenes. Planning scenes. Quiet character moments between allies.
Even here, they shouldn’t agree on everything. They can agree on the goal and disagree on the method. They can agree on the plan and disagree on what it costs. They can agree on the feeling and disagree on what to do about it.
Total agreement is not a scene. It’s a hug. Hugs are lovely in life and boring on the page unless something is underneath them.
If you find yourself writing a scene where two characters genuinely, fully agree on everything, ask whether the scene needs to exist at all, or whether it can be summarised in a single line and cut.
How do you spot fake dialogue in your own draft?
Read the scene out loud. Not in your head, out loud, with your actual voice.
Listen for the moments where one character asks a question and the other character just answers it. Cleanly. Fully. Politely. That’s your fake dialogue. Real people deflect, misunderstand, change the subject, answer a different question, or refuse to answer at all.
Also look for scenes where every line moves the plot forward. That could be an issue. Some lines should move the plot sideways. Some should move it backwards. Some should just sit there being about something the character can’t quite say.
The test is simple. If you deleted the dialogue tags, could a reader still tell which character was speaking? If both voices sound like the narrator wearing different hats, both characters are agreeing on the scene, and both are agreeing with you.
What’s the fastest fix mid-draft?
Pick one character in the scene and give them something else on their mind. Not a bigger something. A smaller one. Something they’d rather be dealing with than the topic on the table.
Maybe they’re hungry. Maybe they’re worried about a text message they haven’t answered. Maybe they’re still angry about a conversation from three chapters ago that never got resolved. Let that pressure leak into their lines.
Suddenly they’re not fully present. They’re half in the scene, half somewhere else. That’s how real people talk, and it’s the fastest way to break the cooperative-actors problem without rewriting the whole beat.
Frequently asked questions
How do I know if my dialogue sounds fake?
Read it aloud. If every question gets a clean answer and every line advances the plot, it’s probably fake. Real conversations are full of deflection, misunderstanding, and half-answers. If your characters sound like they’re taking turns on a podcast, they’re agreeing on the scene.
Doesn’t disagreement in every scene get exhausting to read?
Disagreement about the scene itself is not the same as arguing. Two characters can be warm, loving, and completely aligned on the surface while wanting different things underneath. The tension is subtle, but real. That’s what keeps a reader leaning in without feeling battered.
What if my plot needs two characters to exchange specific information?
They can still exchange it. Just make one of them reluctant, distracted, or trying to steer the conversation somewhere else while the information comes out. Information delivered against resistance is drama. Information delivered cleanly is, well, information.
How is this different from characters just having conflict?
Conflict is usually understood as characters fighting about something they both acknowledge. This is subtler. Both characters can be perfectly polite and still be in completely different conversations. That mismatch is what makes dialogue feel alive even when nobody raises their voice.
Should I plan this out before drafting or fix it in revision?
Both work. Some writers set the mismatched goals before drafting the scene. Others draft cooperatively, spot the flat dialogue in revision, and reassign one character a private agenda on the second pass. Whichever suits your process, the fix is the same: stop letting your characters read the outline.