The Exploration Process
There’s a story I’d like to share.
It’s a story of little sleep as many of these kind are. How do you produce something you have the tools for but not the knowledge? What about in a no fail scenario? One where any solution you come up with, in its final form, has to be 100%? You don’t sleep, you try a ton of different things, and keep your friends on speed dial.
Before the story I’ll get to the meat of the task: the steps to drive to get as close as you can to that 100% in the least amount of time.
1) Set a Hard Deadline
- A week or two is a good amount of time if not too long. < 2 days per revision. This will keep you honest and hungry.
2) Attack the task or portion of the project where you have ZERO idea what you’re doing.
- You can make up ground in places where your knowledge starts on a higher platform.
- It’s very possible that this task will be the part that causes the project to fail. At the very least you won’t see the problems coming. Better to give yourself the maximum amount of time to fall on your face and figure it out.
3) Use what’s already been done to support your driving into the pit of zero knowledge.
- Pull together and cut up fully manufactured plastic cases, repurpose electronics, anything that will be adequate to stabilize the rest of the project.
- You have experience in these parts. Use it to get to 80% quickly.
4) Make some things, break some things, it’s not going to work the first, second, third, or tenth time.
- Get it 85% functionally close to where you need it to be.
- If you’re using CNC anything, or precision anything at this point, you’re doing it wrong.
5) Find the problems and fix them.
- Google it, phone a friend, get to the bottom of why it’s not doing exactly what you want.
- It might be close, but don’t settle for close. Get the zero knowledge portion as close to exact as possible.
6) Dial it in.
- That last 15% is probably the stuff you actually have a basis of prior knowledge on. Great! Now you get to dial the project in.
- What’s wrong? Got it? How do you fix it? Need to redesign the whole thing? Ok, start working.
- Need to switch materials? Parts? Plans? Vendors? Do it. Be glad to move on from your first choice and onto something better. You’re now in search of precision.
7) Get one perfect.
- Got one working? 99%? Woooo! Now it’s time to pull out the CNC and get that ten thousandth tolerance you’ve been dreaming of.
9) Scale.
- You now have a manufacturing package put together of plans and everything you’d need to make 1000 of them.
- Hit the run button again. Make 20 of them. Hope nothing breaks.
8) Drink a beer and sleep.
- If done right, it’s been around a week or so of sleepless nights.
- You know way too much about something new and it needs to marinate in your brain.
- Sleep
Notes:
- This is not meant to yield a product which you can manufacture at tens or hundreds of thousands of units. It is meant to get the a base level of knowledge in an unfamiliar area and execute to a high degree of precision in small quantities.
- Like any good process outline, we need an example. To ground any process throughly in reality it means a human connection - emotions and thoughts. So how about the story now?