If a first act doesn’t set up the goal and the stakes, it’s not doing what a first act should do.

Most of my advice stems from the premise test:

An <ADJECTIVE> <PROTAGONIST TYPE> must <GOAL> or else <STAKES>. They do this by <DOING> and learns <THEME>.

Most movies break down into some form of this. It may follow a group, not a single protagonist (NASHVILLE), the stakes may be low, purely emotional, or metaphorical (BEFORE SUNRISE, MY DINNER WITH ANDRE LAST YEAR AT MARIENBAD), but they will be there and deeply felt.

This is where someone usually calls me out as a hack peddling a formula. This used to bother me, it doesn’t any more. This is just how stories work, some exceptions exist, but you have to dig deep to find them.

Sentences have adjectives that modify nouns that do verbs. Stories have protagonists that pursue goals or else stakes.

This need not be taken as a universal truth, it’s just a convenient and simple way of analyzing a story.


This must be set up in the first 25% of a script. No one need ever say “I’m an alcoholic lawyer who must win this case or lose my soul” (THE VERDICT) but the first act should communicate that as clearly as if they had.

This is where 60% of stories fall apart. The stories fail to set up rooting interest, I’m not clear on what’s going to be pursued, so the middle loses urgency.

Goals and stakes allow us to follow a story. If we like a protagonist and no what he wants, then we can easily track the second act by how each sequence gets him closer to or further from the goal. If we can’t, then it’s really hard to make each individual scene feel necessary and urgent.

Exceptions apply, but this is a generally useful rule. Make sure you first act sets up a rooting interest, if it doesn’t, it’s going to be hard to keep things clear and sustain an idea over the next 70 or so pages.

If structure seems confusing think in these terms: setup/explore/resolve. The exploration is the money part.

The three act structure tends to trip people up. People are either way too into it, or they’re way too dismissive of it. While it’s true that many professional writers don’t set out to neatly color within the lines as they’re writing their work, it’s also true that the three act structure is a useful teaching tool for people who are looking for something, anything to hang an understanding off of as they’re starting out.

This is usually where someone says, “Why waste time learning a technique that real writers don’t use?” This is absolutist and unimaginative. We learn things for many reasons, and the path to understanding often involves learning a lot of indirect stuff. This is an argument that never resolves itself and I don’t feel like having it here.

Besides, most of the terms in a three act structure are just a matter of labeling things that people do naturally. It’s like grammar, when we talk, we’re not conscious of our verbs, adverbs or adjectives. We just make them. You don’t need to know what a gerund is to use one, but if you look at language through the lens of grammar, a gerund will always be a gerund.

Three act structure says that a script can be divided into three basic parts:




This is simple, but unhelpfully vague. Modern three act structure attempts to clarify things by adding a plethora of obligatory points:










I think it’s easiest to look at it like this:


The setup is about 25%, the exploration is about 50%, the resolution is about 25%. You might notice that this sounds a lot like three act structure. That’s because it is. Both are based on the same observation: stories exist to entertain, and we should spend most of the time in the part that’s entertaining.

Going to a restaurant: The setup is getting your table, the resolution is paying the check, the exploration is the actual meal. The meal is the reason we go to the restaurant in the first place.

Sex: The setup is setting plans, the exploration is the foreplay/sex, the resolution is the cuddling (at least from the guy’s perspective) – I’m sure a lady writer would have a usefully different point of view on all this.

Most people are good at setting up, not great at exploring. That’s because exploring is the hard part. Anyone can go out on stage with a violin and a tuxedo, anyone can take a bow – very few can do what’s in between, play a three hour program of virtuoso violin concertos.

Exploration (v): Actively presenting what’s fraught or fun about an idea.

Alternate terms: the money part, the magic, the raison d’etre, conceptual specificity.

All scripts exist to entertain, to transport, to engage emotionally. Genre and concept give clues as to how that entertainment will make us feel, and what things will be used to create that entertainment:

BACK TO THE FUTURE: Genre: Comedy/Adventure. What’s entertaining: The humorous ironies that unfold when a teen meets his parents as peers, with a little bit of skateboard chasing and fighting thrown in.

THE MATRIX: Genre: Action: What’s entertaining? Imagine gunfights, chases and brawls with karate superpowers.

MY DINNER WITH ANDRE: Genre: Drama. What’s entertaining? The depth of the writing and the quality of the acting.

The exploration needs to be 50% of the script – MINIMUM.

Obviously, the setup and the resolution also need to be entertaining as hell, but if you can’t reliably do it in the middle, you’ll have problems doing it anywhere.Exploration should be the bulk of the script, because the Exploration is the most entertaining part and the script exists to entertain.

If exploration is awesome, why not do it all the time? Because setup gives us FRAME OF REFERENCE.

Some ideas are more complex than others. MY DINNER WITH ANDRE takes place at a restaurant. People instinctively get that. So everything that happens feels like an organic progression from a simple start. They don’t need to setup much, beyond the backstory and nature of the friendship.

GROUNDHOG DAY requires a bit more setup. We see Phil being his perfectly assholish self, we see the nature of the day that will be repeated, and we see him quickly figure out the rules of the time loop. After that, we’re off to the races.

Back To The Future and THE MATRIX have incredibly convoluted setups, but we accept them because they enable awesome stuff. Given that their plots revolve around something that is essentially magic, the rules of the magic have to be explained for the human actions to have any weight.

All this sounds good on the surface, but the next logical question is, how do we get there? How do we know what other people will find interesting? The unhelpful answer is, you just kind of know…

The more nuanced answer is, you have to calibrate it against the feedback you receive. This doesn’t mean chasing the market or being desperate to please. It’s about reading the room and knowing the right time to say the right thing. For instance, I find Dungeons and Dragons incredibly interesting. Most of the women I’ve dated have not. This observation resists absolute advice. It’s not that I should always talk about D&D or never talk about it. There are better times than others and the more aware of that I am, the more honeys I’m going to slay.

In closing:

  1. We all of an idea of what is entertaining to us and to others.
  2. The more accurate this idea is, the more successful we’re going to be at communicating.
  3. If we’re putting in effort, but not getting the results we want (which, let’s face it, is most of us) then we should analyze our intial assumptions and literally spell them out.
  4. We should then match our assumptions to our writing and see if it’s literally doing what we set out to do. If it isn’t the flaw is in how we communicated it. If it is, the flaw is in us, and we should work to fix it.





The coda is the little wrap up after the action.

Q: Matt, you sometimes use the word “coda.” Can you explain it?

A:  Coda (noun)

– the concluding passage of a piece or movement, typically forming an addition to the basic structure.

– the concluding section of a dance, especially of a pas de deux, or the finale of a ballet in which the dancers parade before the audience.

– a concluding event, remark, or section

I had to look up the etymology. I always thought it was related to codify, but it’s actually from Latin cauda “tail of an animal.”

Example: If the stakes of the movie are a guy getting his daughter back from terrorists, the climax will be the final fight. The coda will be the happy little scene at the end where the guy and his daughter finally go to Paris or something.

Advice: Many movies wrap up the action too quickly, so their third acts have like 15 pages of coda. Try to keep it around 4 pages, if that.



Storyboard (noun): a sequence of drawings, typically with some directions and dialogue, representing the shots planned for a movie or television production..

This is a storyboard for Toy Story. I like it, because it breaks a scene into shots. If you looked at this on an editing computer, each shot would have movement and motion, but they each represent one camera setup or “take.”

This is useful for screenwriters because it shows an intermediary step between screenplays and movies. Any scene, no matter how complicated, can be broken into a series of index card sized key images. If it can’t, it’s unphotographable, because that’s what the normal process of principal photography exists to turn your words into.

When you’re writing a screenplay, make sure you’re painting a picture. You don’t have to be dogmatic (i.e. every sentence is a new image!) but the quality of your prose should paint vivid, sequential pictures in the minds eye. Storyboards get to use actual pictures, your script should translate those pictures into robust blocks of text that communicate image using the power of the written word.

Screenwriting is about two things: imagining and then communicating. The problem is that imagination doesn’t work like people think it does.

There’s a romantic idea of imagination being like a holodeck, or a magical process. Imagination is actually incredibly limited, because movie making is incredibly limited. Take Star Trek TNG – an entire universe is implied, but most of it was contained in a few sets on a sound stage in Paramount.

Imagination is like a keyhole. We can only see a piece of that world at a time, just like we can only see a small piece of our world at a time. Think of the extruder head of a 3D printer. Things are developed layer by layer, line by line.

This is a lot like how screenplays work – no matter how vividly we’ve imagined something, we can only communicate it to the readers line by line.

I like to run clients through this exercise:

 ME: I want you to imagine a man running for a bus?
 GUY: Okay, got it. 
 ME: What is he wearing?

This usually gives half of them a pause because they weren’t actually imagining the guy real well. They weren’t pretending, they were pretending to pretend. The other half half will dutifully record what the person is wearing. The vast majority of them will say “a suit” because it’s the easier to describe.

ME: What color is his tie? (1)
GUY: Uh... it's red.

This gives almost everyone a pause, because even though they may be picturing a guy with a suit, few people have the visual imagination to summon up an entire suit along with tie and belt. Once they look for the tie, it fills in, and they begin to see it. (2)

So now we have a well-dressed man running for a bus. There are a few avenues to explore: where is he going? Why doesn’t he have a car? Why didn’t he call an Uber? Sometimes I’ll ask these, most times I’ll ask this:

 ME: What are in his pockets?
 GUY: ...Uh...

This always gives people a pause, because honestly, when was the last time you imagined the inside of a character’s pockets? Like the tie in this example, it’s not there unless you look for it. The more specifically you imagine the guy, the easier it is to populate his pockets. (3) Usually I get some version of this answer:

 GUY: Uh, his phone... and a pocket watch (or handkerchief, pen, condom or the like).
 ME: Okay, we already have a problem. He's running for the bus, but he doesn't have his wallet or any money.

Now, this is the world of imagination, if we want him to have a wallet, we just fill it in. But this is a writing exercise, so the lack of the wallet is more interesting. This raises some questions. Where is his wallet? How did he lose it? Does he know he doesn’t have it or not?

GUY: He knows he doesn't have it. But he's going to ask the bus driver nicely, and he'll probably get on.
 ME: Okay, I can buy that. Bus drivers let people slide all the time, and he's plainly desperate. Where is he going? 

And here’s where things get interesting. He could be going to work, he could be on his way to a job interview, a lot of people say he’s going on a date (which leads me to question how he’s going to get around the fact that he has no money). The more urgent the need, the more fraught with dramatic possibility the scene becomes.

But let’s say they remembered to account for the wallet.

ME: So he's got a wallet? What's in it? Credit cards, money, a driver's license?
GUY: Yeah.
ME: So if he has a license, there's stuff on it. What's his name?

This gives everyone a pause, because imagination simply doesn’t work this way. So now they have to look at the blurry, shittily formed license facsimile they’re imagining, look for a name that isn’t going to be there, and then come up with one.

GUY: Roy.. Smith.
ME: What year was Roy born?
GUY: Aw, come on. 
ME: If he has a license, it has his age. How old is he?
GUY: Uh...

This exposes another flaw of imagination, because people tend not to see an actual person, but a vague man shaped blur. Now they’re forced to envision the guy better, and do the math and determine a birth year. If they know anything about horoscopes, I’ll ask them to give him one. Most of my clients don’t.

Then I’ll ask their address. Because they’re not really seeing the license, they won’t know it, so instead, I’ll ask them to say the kind of dwelling they live in, if they have roommates, how they feel about the roommates, etc.

At the end of the exercise, we’ll end up with Roy Smith, age 37, who lives on Orchard Lane in a two bedroom apartment he shares with his roommate Kenny, a lazy stoner. He’s running to a date with a dream girl who he met at Costco, and now he’s desperately hoping that Kenny can meet him downtown and hand over his wallet, which he lost at his girlfriend June’s apartment when they broke up.

We now have character, stakes, a world, and many fraught scenes. We’ve imagined Roy with a deeper clarity and understanding than many beginning writers have for the characters in the feature scripts they’ve spent months developing. And it all started with the simplicity of a man running for a bus.

Try this for yourself, or more interestingly, try it on an unexpecting victim. You’ll be surprised what you come up with.

NOTES (1) Not everyone has the same level of visual imagination. Some people are better at image, some people are stronger at coming up with relationship, emotion, or fraught scenarios. Your average costume designer is going to be way better at this job than your average carpenter.

(2) This exercise becomes much easier if you picture a specific person, like your dad or Harry Potter. Most people don’t, if they’re too good at this, I’ll explicitly ask them to make an original character.

(3) Movie characters rarely have anything in their pockets because it ruins the lines of the clothes that wardrobe has picked out. A few actors like to have some objects to ground their character plus the clout to get their way, but it’s rare.