HomeVideos

Give Me 13 Minutes, I'll Make Your NPCs 10x Better

Now Playing

Give Me 13 Minutes, I'll Make Your NPCs 10x Better

Transcript

374 segments

0:00

Most NPC advice out there focuses on

0:02

voice, personality, and quirks. But

0:05

after many years of playing this game, I

0:07

realized that this is actually backwards

0:09

and leads to NPCs everyone forgets and

0:11

to games that all feel the same. So

0:14

today, I'm going to fix this by giving

0:15

you a five steps method to create

0:17

amazing NPCs. And to make it easier to

0:19

remember, I will use it on a very basic

0:21

character and transform it into one

0:24

worthy of your table. But just before we

0:26

dive into this method, I just need to

0:28

clarify. What do I mean when I say we

0:30

build NPCs backwards? Well, I think

0:32

instinctively we start on the outside.

0:34

We all have lists of names and a few

0:37

default voices, sometimes a short

0:39

backstory. And because our players love

0:41

to surprise us, we use them, put them in

0:43

front of the players and hope it does

0:45

the trick. And sure, sometimes it does

0:48

and it's all you need. But most of the

0:50

time, three sessions later, players

0:52

can't really remember their name. And my

0:54

take is that it's not because you suck

0:56

as a DM, but because you built it in the

0:59

wrong direction. I think we're so scared

1:01

of hurting player agency that a lot of

1:03

us default to superficial NPCs to avoid

1:06

railroading, even when building the more

1:08

important ones. And when we don't, we

1:10

tend to treat NPCs like we're writing a

1:12

character in a novel. And that's a

1:14

problem because those need to be

1:16

interesting in themselves. But an NPC at

1:19

the DND table doesn't. They need to do

1:21

something to your players. This is very

1:23

different and it's what separates a

1:24

character and a force which creates

1:27

pressure which then leads to memorable

1:29

moments at the table. Okay, so let's

1:31

build this NPC. It will make it really

1:33

easy to understand. Let's start with the

1:36

basic version and fix everything that's

1:38

wrong with it as we go. I'm going to

1:40

call him Mever. I think Mever is a

1:42

middle-aged human man. He's a little

1:44

chubby and he is a merchant. He sells

1:46

cloth and has been in town for about 6

1:48

weeks right now. Now, he's pretty much a

1:50

prop with a job, a name, and let's be

1:52

honest, absolutely no reason whatsoever

1:54

for everyone to remember him. But that's

1:57

going to change very soon. So, the first

1:58

thing our NPC is missing is a want. He

2:01

has a job, but this is a game, so we

2:03

need a want to put the NPC in motion. If

2:05

we just give the NPC a job, it will only

2:07

react when the players show up. But a

2:09

want means the NPC will have a reason to

2:12

do stuff, which will make it feel very

2:14

different. An NPC without a want doesn't

2:17

really feel wrong or unbelievable, but

2:19

just kind of empty, which makes them

2:21

easy to forget. Right now, Maver has a

2:24

job. We said he sells cloth. Uh, but if

2:26

we stop there, even if we decide to make

2:28

him a quest giver, here's what's going

2:30

to happen. The players will walk into

2:31

his shop, buy a blanket or something,

2:34

and he will tell them that there has

2:35

been bandit activity on the road, which

2:37

is bad for business, and ask them to

2:40

handle it for some gold. The players

2:42

will go kill the bandits, level up, and

2:44

get paid. And then they will never think

2:46

about him again. Mever will have played

2:48

his part, and there will be nothing left

2:50

to think about. So, let's give him a

2:52

want and see what happens. I think Mever

2:54

wants to recover a journal. His younger

2:57

brother was a novice in the clergy of

2:59

the world's god of light, and he wrote

3:01

down everything he saw in the monastery.

3:03

Unfortunately, the brother died, and the

3:06

journal disappeared. and Mever has been

3:08

looking for 2 years and chose to settle

3:11

in this specific town because he thinks

3:12

it might get him closer to the journal.

3:15

Now the NPC feels completely different.

3:17

The dude is still a merchant. He looks

3:19

the same and everything, but now every

3:21

conversation can have this subtle

3:23

undertone and the same bandit quest can

3:25

become a lot deeper. Maybe he thinks the

3:27

bandits could know something, have the

3:29

journal. We have room to grow. By the

3:32

way, if you've ever struggled with NPCs

3:34

before and find this video helpful,

3:36

please hit the like button with your

3:38

most powerful ability to help it reach

3:40

more people and create better games for

3:42

everyone. But a want is not enough. A

3:45

lot of DM stop there, and it's a shame

3:47

because if you're not careful, a want

3:48

quickly starts to feel a bit like a

3:50

shopping list. It tells you what someone

3:52

is doing, but the why is often still a

3:55

bit fragile. If we want the want to

3:58

trigger some kind of story, we need more

4:00

emotional weight. If the want is a bit

4:02

too clean or logical, the players will

4:04

just solve that and move on, which is

4:06

not great story material. But if we

4:09

stack a wound on the want, it becomes

4:12

much harder to solve the NPC because the

4:15

want stops being just about the thing

4:17

they're chasing and it starts to be

4:18

about trying to prove something or

4:20

become someone. And this opens up a lot

4:23

of role-play opportunities which will

4:26

make it easy for players to feel that

4:28

depth. The wound acts as some kind of

4:30

texture generator. So, how does it work

4:32

with Mever? So far, he just wanted the

4:35

journal because it belonged to his

4:37

brother, which yeah, of course. But if

4:39

we stop there, players will find the

4:41

journal, give it to him, get the gold,

4:43

and that's it. Again, he'll be forgotten

4:46

soon. So, let's get him his very own

4:48

wound. I think Mever was on the road

4:51

with the journal when Bandits attacked

4:53

and he got scared and fled, leaving

4:55

everything behind. His brother was

4:57

already dead, so it's not like he

4:59

sacrificed him or anything, but their

5:01

journal was the one thing he had left

5:03

and he didn't try to fight for it. I

5:05

like this as it would. It's a mix of

5:07

shame and regret. And it changes the

5:09

want quite a lot. Now he's not just

5:11

trying to get some journal back, but

5:13

trying to forget that coward version of

5:15

himself. and players can't solve him

5:17

anymore because giving him the journal

5:19

doesn't really close the wound. Our

5:21

little NPC needs to see himself act and

5:23

in the end the journal is just a symbol

5:25

he's decided would prove his capable.

5:28

It's quite irrational and therefore

5:30

human and more interesting to play out

5:31

on the table. But there is one very

5:33

important thing you need to know about

5:35

wounds which will have an impact on how

5:37

you use them. They don't have to be

5:39

visible. Players might never take the

5:42

conversation in a direction that would

5:44

make them learn that Maverick ran, but

5:47

you as the DM know. And that will color

5:49

how you respond to the player's actions.

5:52

Say the players do decide to help out.

5:54

Maybe he'll be a little too grateful and

5:56

still feel odd because deep down he

5:58

knows he needs the help because he

6:00

failed. This is how you know that your

6:02

wound is a premium wound. All right, so

6:05

this is cool. But if we want this NPC to

6:07

truly be one of those players talk about

6:09

for years, we're still missing

6:11

something. I think the best NPCs are

6:13

never fully legible. What do I mean?

6:15

Well, I think there is always something

6:17

that doesn't 100% add up. And of course,

6:20

I say this and you immediately think,

6:22

"Oh, yeah, of course, a secret." And

6:25

yeah, kind of. The problem with most

6:27

secrets is they tend to be a little

6:29

passive, like a dark past, for example.

6:31

It's hard to make it have an impact on

6:33

the game. But I think that active lies

6:35

make NPCs kind of feel like puzzles and

6:38

that's way more interesting to play. I

6:40

think your NPCs need to have something

6:42

they're currently doing to protect what

6:44

they want. It's all about it having an

6:47

impact on their behavior. And this

6:48

matters a lot because if you give them

6:50

something like this, you don't have to

6:52

lure dump it on your players and they

6:54

can actually see the secret in action

6:56

before getting to the bottom of it. that

6:58

will trigger some actions on their part

7:01

and therefore make the NPC feel way more

7:04

memorable. And so far, the way we set up

7:06

our example NPC does not really allow us

7:09

to do this. He has the want and the

7:10

wound, but they're kind of separate from

7:12

his merchant identity, which will make

7:14

it hard to observe. So, let's give him a

7:17

cover. I think the way we do it is we

7:19

tie those together like this. His trade

7:21

is completely real and he sells the best

7:23

blankets in the region. But he has

7:25

recently changed his suppliers just to

7:28

justify setting up shop in this specific

7:30

town because he knows the bandits are

7:32

closed and have the journal. And now you

7:35

can have the players hearing rumors from

7:37

those suppliers at the tavern saying

7:39

that they don't actually understand why

7:40

they ditched their previous ones which

7:42

were better or guards saying they

7:44

couldn't refuse him entry but found it

7:47

odd that they'd come here because he

7:49

came from a wealthier region. It makes

7:52

coming up with a hook much easier and

7:54

gives more texture to the NPC. We're

7:56

looking good so far, but we still have

7:58

two more layers to go through before

8:00

we're done with Maver. And just before

8:02

we see what we're still missing, I just

8:04

wanted to let you know that if you find

8:06

this stuff useful, but don't know where

8:08

to start, I actually made something that

8:10

might interest you. It's called the

8:12

living town. It's a collection of 15

8:14

interconnected NPCs with wands, secrets,

8:17

and store hooks baked in that you can

8:19

just drag and drop into your campaign.

8:21

They give you a great starting point and

8:22

will make your next session even better.

8:25

So, just check out the link in the

8:27

description. All right, so we have the

8:29

core of our NPC, but when the players

8:32

aren't there, what is he doing? I feel

8:34

like most of the time NPCs just

8:36

basically stop existing when the

8:37

characters aren't interacting with them.

8:39

And that's really not like how people

8:42

work in real life. And of course, NPCs

8:45

aren't really people, but this

8:46

contributes to this we don't care

8:48

feeling we've been trying to solve

8:50

throughout this video. I think NPCs,

8:52

just like real people, should always be

8:54

in the middle of something. You need

8:56

movement to avoid your game feeling like

8:59

a film set. So, the way you fix this is

9:02

you have your NPCs already doing

9:04

something when they encounter the

9:06

characters. They could be talking to

9:07

someone, organizing things, writing

9:10

something. There's a million ways to do

9:12

it. But this has a massive impact

9:14

because now when the players walk in,

9:16

they don't feel like they're just

9:17

starting a quest, but discovering an

9:20

area of the world that already had its

9:22

own independent life. So, if we go back

9:24

to our NPC right now, there are things

9:27

under the hood, but they're just sitting

9:29

there waiting to be discovered. So,

9:31

let's give him a bit of momentum. What

9:34

if about 2 weeks ago, Maverick paid

9:36

someone to try and locate the bandit

9:38

camp where he suspects the journal might

9:40

be, and he hasn't heard from the guy

9:42

since? Is he dead? Maybe. Or maybe he

9:45

found the journal, realized what he had,

9:46

and is currently wondering if he should

9:48

take it to the clergy to maybe get an

9:50

even bigger reward. And this is making

9:53

me never nervous. So now when the

9:55

players walk into the shop, they find

9:57

him pacing like a maniac, maybe talking

9:59

to himself, slightly agitated, and he

10:02

stops immediately when he sees them. And

10:04

what I really like about this is it

10:06

didn't just make the character more

10:08

interesting, but it also helped build

10:09

the world. Now we have another NPC, the

10:12

informant, and he could turn out to be

10:14

100 different things. This force idea we

10:17

mentioned at the start is really

10:19

starting to take shape, but something is

10:21

still missing. How do we make sure that

10:23

we're not just building a beautiful

10:25

railroad? Because when you keep adding

10:28

layers to NPCs, they become more

10:30

defined, which means you risk more to

10:34

either see the players do something you

10:35

never planned and become frustrated, or

10:38

to force them down a specific path,

10:40

which will be frustrating for them. The

10:42

way you avoid this and keep everyone at

10:45

the table happy is you let the players

10:47

build an ignition point. What is that?

10:50

Well, let's go back to that aggressively

10:52

pacing moment. When the players enter

10:54

the shop and see the NPC just furiously

10:57

walking by himself, they can react in

10:59

different ways. Maybe they will become

11:01

suspicious or have compassion or find

11:04

him funny. Your job is to pay close

11:06

attention to their reactions in this

11:08

moment and to make it shape where the

11:10

story goes from there. For example, say

11:12

the players find the pacing quite

11:14

intense considering the shop is empty

11:16

and their brain goes to this looks like

11:18

scheming. And now the way they talk to

11:20

the NPC is a bit aggressive. Well, he

11:22

can give them a few clues, lead to the

11:24

quest, but decide that actually Meford

11:26

wants the journal to have some kind of

11:28

insurance against the clergy and

11:30

blackmailed them. If the players see a

11:32

lost man trying to put on a brave face

11:34

when they enter and act accordingly, you

11:36

can decide that Mever is being taken

11:37

advantage of by the informant who wants

11:40

to betray him and that the informant is

11:42

going to become some kind of lower level

11:44

villain. The point is the reaction of

11:47

the players is insight into where the

11:49

table is at right now. And it gives you

11:51

ammunition to then take the story in a

11:54

direction that will feel fun for them

11:55

and for you. Because even though you did

11:58

plan a few things ahead, you get to

11:59

modify them on the fly and be surprised

12:02

by the collaborative aspect of it all.

12:04

So, are we good then? Well, let's go

12:06

back to the beginning. Our NPC is still

12:08

a slightly awkward middle-aged human man

12:10

who sells blankets and whatnot. So, the

12:13

shape is still the same, but what's

12:15

underneath is a lot deeper. He wants

12:18

something that's costly. He's running

12:20

from something he can't outrun. He has

12:23

an active secret. He's in motion. And we

12:25

have several doors opened. So, we know

12:27

that no matter what the players do,

12:29

we'll be able to tell a cool story with

12:31

them. And just like that, we went from

12:33

boring NPC nobody will ever remember to

12:36

force, which will generate story and

12:38

make the world feel alive. Am I saying

12:41

you absolutely need all five layers for

12:44

every single NPC ever? Of course not.

12:47

You can definitely pick a few of these

12:48

elements depending on the type of NPC,

12:51

it's role in the story, etc. Maybe you

12:53

start with just a name cuz your players

12:55

did something you didn't plan. A cool

12:57

moment happens as a result and before

12:59

your next session, you decide to use

13:00

these steps to give it more depth and

13:03

make them have a role in the story. You

13:05

can definitely do that. The result will

13:07

feel even more organic and memorable.

13:10

What matters is the direction. You start

13:12

with a wants and build from there. But

13:15

now that we know how to make amazing

13:17

NPCs, we have a problem. Because even if

13:20

your NPC has a want, how do we make sure

13:23

the world around them works as well as

13:25

they do? I believe the easiest way to

13:27

make sure it is the case is to build the

13:30

world and that you should 100% make your

13:32

own to become a better DM. So, click on

13:35

this video next to find out how.

Interactive Summary

The video outlines a five-step method for creating memorable NPCs in D&D, countering traditional advice that often leads to forgettable characters. The core idea is to build NPCs from the inside out, focusing on elements that drive player engagement and narrative. The method includes giving an NPC a 'want' (a driving goal), a 'wound' (emotional weight for that want), a 'cover' (an active secret or lie influencing behavior), 'momentum' (being actively involved in something off-screen), and an 'ignition point' (allowing player reactions to shape the story's direction). This approach transforms basic NPCs into dynamic forces that make the game world feel alive and stories more engaging.

Suggested questions

8 ready-made prompts