DDeep Dive with Ali Abdaal
← All frameworks
InnovationBrandon Sanderson

The Chef, Not the Cook: Deconstruct-and-Rebuild Originality

Difficulty
Moderate
Time to result
~ongoing to results
Steps
6
Confidence

Sanderson teaches his students that creators fall into two modes. A cook follows a recipe and is helpless when it fails; a chef understands why each element works and can diagnose, adapt and invent. His model of originality follows from a claim about how human creativity actually operates: we recombine rather than invent from nothing — we put a horn on a horse. So instead of chasing wholesale novelty, you train yourself to take work you love, break it down to the fundamental reason you love it, and rebuild that core element inside something new. Done well, the result feels wholly original even though its structural DNA is borrowed, and early derivative work stops being a source of fear.

Origin

Extracted from Deep Dive with Ali Abdaal

How to run it

  1. 1

    Accept that creativity is recombination

    Drop the premise that original work means inventing something with no antecedents. Human imagination remixes known parts; it cannot picture a colour it has never seen. Your job is novel combination and novel framing, not novel raw material.

    Pro tip Reframing this early removes most of the paralysis that stops creators from starting.

  2. 2

    Pick work you genuinely love, not work you admire

    The method runs on emotional response as its raw signal, so choose the books, films, videos or games that actually move you rather than the ones you think you should study. If you feel nothing, there is nothing to decompose.

    Watch out Analysing prestige work you don't enjoy produces imitation without insight.

  3. 3

    Break it down to why you love it, at a fundamental level

    Interrogate the piece until you can name the mechanism, not the surface. Not 'I like heist movies' but 'I like the promise of a plan, the reveal that the plan was a lie, and the crew's competence under pressure.' This is what separates the chef from the cook.

    Pro tip Ask what specific beat you would miss most if it were removed. That is usually the core.

  4. 4

    Look for the same structure in unrelated genres

    Compare across categories to see the pattern underneath. Sanderson points out that a buddy cop movie and a romantic comedy often share the same plot structure. Cross-genre matching proves you have found structure rather than trappings.

    Pro tip Keep a running list of structures you've identified; they become a reusable toolkit.

  5. 5

    Rebuild the core element inside a new container

    Take the isolated element and construct something of your own around it. Inception follows the heist beats faithfully but transplants them into a setting no heist had used, so it reads as wholly original while being structurally familiar.

    Pro tip Change the setting, stakes or medium, not the beats — the beats are why people love it.

    Watch out Swapping the surface while keeping the specific characters and imagery is copying, not rebuilding.

  6. 6

    Let the early work be derivative on purpose

    First attempts will feel derivative, which is why first novels usually stay unpublished. Voice emerges through the attempt to sound like the people you admire, so keep producing rather than withholding work until it feels sufficiently original.

    Watch out Waiting for originality before publishing is the most common way creators never publish.

In the wild

Inception as a rebuilt heist

Sanderson uses Christopher Nolan's Inception as the clean demonstration of the method. The Nolans clearly love heist films, broke the form down to what actually makes a heist satisfying, and then constructed their own version around that core. The film hits the heist beats accurately and yet feels unlike any heist audiences had seen, because the element that was borrowed was the underlying structure rather than the setting, characters or imagery.

A film that reads as wholly original while running on a completely conventional structural chassis.

The Bill Wurtz lineage on YouTube

Sanderson traces a YouTube example: Bill Wurtz released a history of Japan video, it was wildly successful and innovative, and then he moved on to making music. Years later entire channels exist built around comedic retellings of factual events that owe their DNA to that one video. Several creators openly acknowledge watching it and deciding that since nobody else was making these, they would. Each channel still feels different and original, and several have hundreds of thousands of subscribers.

A whole content category grew from one deconstructed influence, with each descendant developing its own voice.

Ali's YouTube coaching version

Ali applies the same loop when teaching creators to grow on YouTube: look at channels whose videos you actually enjoy, work out specifically what it is about them you like, then think about how to incorporate that into your own videos rather than worrying about originality at the outset. His observation is that the pursuit of imitation is itself the path to a distinct voice — in trying to be like someone you admire, you end up finding your own way of doing it.

Creators ship consistently early and differentiate later, instead of stalling on the search for a unique angle.

Common mistakes

Chasing wholesale novelty

Trying to invent something with no antecedents fights how human imagination works. Sanderson's point is that we put a horn on a horse; we don't conjure creatures or colours from nothing. Creators who hold out for a totally unprecedented idea produce nothing while the recombiners ship.

Copying the surface, not the structure

If you take the setting, imagery or characters instead of the underlying mechanism, you get imitation that reads as derivative. The chef's move is to identify why the thing works, then rebuild that reason inside a container of your own.

Fearing plagiarism and derivativeness

Sanderson isn't afraid of being plagiarised, because authors develop their own voices — give two writers the same concept and you get wildly different books. Guarding ideas and worrying about being derivative in early work wastes energy that should go into production.

From the transcript

the way that human creativity works is recombination we remix that's what we're really good at

Brandon Sanderson · 18:30

train yourself to look at something you love and break it down to why you love it on a fundamental level and then rebuild it…

Brandon Sanderson · 00:00

if you can learn to to boil it down to what you love and take that core element and build something new around it you…

Brandon Sanderson · 20:30

From the episode

Bonus Episode: Originality, Writing and Career Fulfilment with Brandon Sanderson

Brandon Sanderson