Working in high tech helped us to start our food project.

Food is important to me. From the place I was born to the place I grew up and at every place I’ve worked or lived, food has played a central role. Food connects people. No matter it is considered luxury or commodity, in almost every civilization, food has a very unique place.

And I like food so much, I started cooking when I was about 10. Prior to that, I just watched my parents cooking.

It was amazing.

Looking at the transformation of the ingredients, how the colors change with temperature and time, just like magic. Smelling new flavors, created by the imaginations of men, and rediscovering them, rushing from a place called nostalgia. Touching the texture, playing with it. And listening to it. This little music under our teeth, vibrating, cracking, and guiding our taste.

Food is one of these unique experiences where every sense plays a role.

But like many, I didn’t take this path. None of us did. Instead, I started a high tech company and my friends went working for other digital companies. Since then, we lived with computers.

We never stopped cooking. We never stopped baking. We never stopped dreaming a day we’d become Chefs.

The initial idea came after the realization that very few people actually bake at home. Many people say they like to bake or would like to bake, but they don’t. And when they do, it’s always the same things: brownies, cookies, cupcakes or other simple things. And the simpler it is, the better. Some of them put frozen dough in the micro-wave and call it baking. Well…

What about the more complex pastries? We come from Paris, France and we grew up eating “le millefeuille”, “l’Opéra” or “le Paris-Brest”. Why don’t people make them at home? What’s hard in making “le macaron Parisien” that everybody loves? What’s complicated in baking “la tarte aux framboises”, a classic!

Because execution is hard.

Here’s the test: let’s say you want to make an apple pie. Nothing hard, right? You can probably find all the recipes in the world on the internet. And if you want to play fancy, buy a cookbook from a great chef. Then, go to the closest grocery store to get the ingredients. And finally, follow the recipe.

Right?

Here’s the reality: first, we spend an hour or two deciding which recipe to follow. There are so many choices, especially when it comes to something very “classic”. Yup, that’s the irony: the more “classic” it is, the more variations it has.

Then, once decided, you got to go to the grocery store and hope to find everything. Even if you happen to find everything in one place, you have probably spend about another 2 hours just looking for the right brand and making the queue to pay.

Now, back at home, it is time to understand what’s the recipe really says, clean, measure and scale everything. Plus, you have to make sure you got all the special utensils that are required.

Basically, before any baking, you have already spent half a day preparing everything start.

And finally, comes the baking.

One of the main issues with cookbooks and recipes is they are designed to be “generic”. Of course, the writer has no idea who the users are and has no idea what brands, what ingredients, what tools the users are going to use. This leads to the use of some jargons that hardly anybody understands (“monter une crème”, “déglacer” or “réduire un jus”) and to some confusions (use “flours”… ok, but there are more than 10 different “baking flours” just at Whole Foods… which one should be really used?).

The similar problem exists in high tech. How many people would like to start coding? Developing an app for example? How many time do you hear somebody saying: “I got a great idea for an app” but doesn’t do anything?

Why?

Again: execution is hard. Hard to find the “language”, hard to go look for the “best tutorial”, hard to start learning, hard to understand what’s in the tutorial, hard to re-learn some math we forgot or never learnt at school because we were taking a nap or talking to the cute girl/boy next to us, hard to actually code something etc.

But over the years, because software development has become so key and speed is important, a solution has been found: SDK. Software Development Kit.

The concept is easy: what if all the building blocks are already pre-established, well designed and we only teach the users how to move these blocks and put them together?

Think of LEGO: when you buy a box of LEGO, let’s say the R2D2, you have everything you need to build it. It comes with all the pieces, designed to fit together so you don’t need to cut any plastic yourself. You just need to put them together. How? The box contains a manual. Dedicated manual that only uses the exact pieces that are contained in the box. And if you follow all the steps, you’ll end up having the same R2D2 that is pictured on the box.

No need to have any degree in design or in construction to build a house. No need to be a rocket scientist to build a Spaceship. No need to be an engineer at Industrial Automaton to make the R2D2.

Better: if you are creative, you can also build without any manual and create your own thing.

SDKs work the same. Just the implementation (putting blocks together) is “a bit” more complex.

In early 2015, my colleagues and I designed and developed the SDK for our software company. I learnt about how it is done and all the challenges.

It was then obvious to us that the best way to teach baking and democratize this passion is to create a Pastry Development Kit.

There are few things we immediately defined as our key objectives:

First, our objective is to make a kit that helps the users, any user, to make real, beautiful, pastries. Not just cookies. But pastries that can be eaten at starred-restaurants and great professional bakeries in Paris.

Also, our mission in the long term is to help people to fall in love with baking, or cooking in general. To achieve this, the execution has to be fun. We need to find a way to transform the struggle and the grind into something meaningful and valuable. We have to find a way to make people love the experience of baking and not just the outcome of baking.

When we started thinking about how to do it, we recalled some lessons learnt designing and using different SDKs.

To make a great development kit, the documentation and the “parts” need to be designed in harmony. Our user’s manual can’t be just “another generic recipe” and the “parts” can’t be “just ingredients anybody can buy at Whole Foods”.

Documentation was a big issue: people don’t read it. One of the best way to make great documentation is to have it in the code itself and make it interactive.

We designed our first documentation without any text. Only pictures. The idea is anybody would understand it, even without knowing how to read. So instead of having one big picture and the explanation next to it, like almost every recipe, we had break down every step into different action, just like comics.

The ingredients and utensils are specifically designed to guarantee the success if all the steps are followed. The documentation uses exactly the same picture and depicts exactly the same ingredients and utensils that are in the box.

We also chose to draw everything instead of taking pictures. There are 3 reasons behind this decision.

For us, pastry has a part of dream and should be warm. Because we aren’t good in photography, it is hard to recreate this sensation.

The second reason behind this artistic choice is we want things to be simple. We want the users to focus on the essentials, the high level, instead of losing focus on the small details.

Finally, for us, PastryNow is very similar to “Ratatouille”, the Disney animation. We want our users to be a Pastry Chef. We want our users to be the hero of this great culinary adventure. Choosing to represent everything in a cartoon fashion helps to develop the feeling of “being in a game or in an animation”.

Of course, nothing’s that easy. Even when (you think) you know everything, converting ideas into concrete products is a very difficult task.

We spent 6 months developing the entire project (also have to take in account the online store, the website, creating the channels, doing the market research etc.), launched 3 different “Magic Kits”, made over 40 different iterations, spent hundred of hours listening to users’ feedbacks and we finally got something that is close to what we initially envisioned.

What’s next?

Make the kits better. Let more people know about the kits. And repeat these 2 steps faster.

Just like a computer.