When you start learning to cook, you compulsively follow the recipe. You stress over timing and ingredient ratios and whether the carrots are truly and properly chopped or just somewhat sliced.
Over time, you learn where the wiggle room is. A lot of recipes simplify into:
Mix it all together and cook on hot-ish with some oil until squishy and then add some cheese.
Most recipes work like that.
We’re able to use shorthand with food because we know what the sane bounds regarding “squishy” and “some cheese” are. Unless you live in Britain.
With startups, we don’t yet have those bounds.
We lack the intuition to distinguish between a raw, rotten, and yummy conversion rate or user story.
A lot of startup advice culture is at the stage of reading a step-by-step recipe and then saying:
Aha! But on my oven, 6 really means 7!
It disguises itself as creativity, feels like customisation, and is still just imitation.
Constraints comes in two flavours:
- Process constraints which are solved through creativity
- Creative constraints which demand a specific process
Making a tasty sauce to put on a steak with a fridge full of leftovers is a case of the former. You’re going to come up with some sort of crazy sludge and it might be delicious! It’s fun and you end up surprising yourself.
It can also end in a ruined meal (which isn’t so bad, really), so people tend to be overly conservative and have to be forced into it by not having everything they want available.
My favourite course in uni was one of Ian Bogost’s about translating haikus and ancient greek love poems into video games. You can’t use FPS design tropes, which leads to some weird and [potentially] wonderful stuff.
(Incidentally, if you’re the curious sort, here is a food-themed game I once made about The Lovesong of J. Alfred Prufrock. It’s a java applet hosted on an embarrassing snapshot of an old university blog of mine, but such is life)
The second type of constraint is about execution. Your inventive new sauce notwithstanding, if you cook the steak for 30 seconds too long, the whole exercise is fruitless.
Adwords optimisation is pretty much a solved problem. You don’t need to re-invent the idea of funnels. Once you’ve decided that’s what you want to eat, those pursuits are about execution and best practices.
Part of the benefit of following pedantic instructions for a while is that you figure out where you have wiggle room and where you don’t.
Recipes are great. You should use them! And when you find a good one, please share it. I get a load of value from other founders’ play-by-plays.
But recipes are also transitional. Chefs imitate until they can create. And then begins the exciting time — for the chef, for the diners, and for the cooking community at large who can continue to build their on originality.
 In case you were wondering about bagels, you put flour and a spoonful each of yeast, salt, sugar, and oil in a bowl. Mix it and mush it together while pouring in water until it’s a big ball. Boil a pot of water, turn on the oven, make toroids, boil them for a minute each, dip them into whatever you want stuck on the top, and put them in the oven until brownish.
As weird as it seems, don’t forget to boil the dough. An unboiled bagel is just a roll with a hole.