A physicist's guide to learning hard things (and using the scientific method)

[lowtek]

So I was responding to a thread, and realized that one problem I see with some folks, newbies in particular, is that they lack a systematic methodology for teaching oneself. This isn’t surprising, as it’s not really a skill that we’re taught early in life. Realizing this, I figured I would share what I have learned about learning difficult things, over the years.

Of course, the scripted dogma is that we must go to school for 4 years, rack up a mountain of debt, and hope that we'll have marketable skills at the end. For STEM fields, this actually isn't a bad bet, but since most people don't major in useful things, it turns out disastrously for the majority.

I actually see some parallels between that scripted dogma and some of the posts here. Some folks will do extended learning challenges, work through dozens of tutorials, or seek out mentors to hold their hand... Perhaps it's the case that some people really need these constructs and arrangements, but I believe that there exists a better way to teach oneself.

Full disclosure: I guzzled the scripted dogma for the first thirty years of my life. I loved undergrad so much that I did a 5th year as a victory lap. Not willing to enter the real world, I then went on to a masters program and ultimately a PhD. However, from the ashes of these potentially ruinous decisions came an invaluable skillset that benefits me to this day: the ability to learn new things with minimal guidance.

Much like we have the CENTS framework for evaluating business opportunities, I believe there exists a mental framework for teaching oneself new skills.

What does it mean to learn something, really? I argue that learning means internalizing information in a way that allows you to apply it to new problems that are only marginally similar to previously encountered problems. After all, if you only know how to solve one problem, you’re not really a problem solver, you’re a robot.

Central to learning is having a solid mental model of cause and effect relationships. This is necessary because the world we live in is causal. To illustrate, let’s consider a simple example.

Let’s suppose you’re selling a widget on your website. For the sake of argument, we’ll assume that the widget is something people actually want. You’ve got some traffic, but only 1% of it is converting. You want to increase the conversion rate, because you have big baller dreams of lambos and lavish nights out in the big city. Problem is that with that 1% conversion rate you’re barely making burger flipper money, so hyper cars are out of the question, let alone hiring someone to fix the landing page. It’s on you to solve this problem. How to do it?

Most people will approach this problem by reading blog posts, books, or listening to podcasts on conversion rates. They will think they’re taking action, and assimilating knowledge, but in reality they’re wasting time. There are two reasons for this:

1) This delays the feedback loop between taking action and receiving results, thus obscuring the relationship between action and results.
2) It’s often event focused rather than process focused. Consuming prepackaged information often obscures the why of the solution.

The correct approach to learning how to increase conversion rates is this.

Clearly state the problem; the more precise and detailed the better.
Make a guess about what things you can control contribute to the problem
Formulate a plan for iteratively testing each of these possible root causes. Test most likely cause first, if possible.
If necessary, seek out information that tells you how to address each possible root cause.
Execute the plan.
Repeat until solved.

So if your conversion rate is only 1%, you could guess that the following things contribute to that low rate:
1) the page loads so slowly that people leave before it finishes
2) The color scheme makes peoples’ eyes bleed, so they can’t find the buy button
3) The copy reads like it’s written by an overseas scammer
4) The perceived risk is too high

These are just a subset of the possible reasons, and indeed a combination of them could be the problem.

We’ve already stated the problem, so that’s good.

Our analytics indicate loading speed could be an issue, and indeed any of the other 3 could be the problem so we have no choice but to test them all.

Brute force it is.

We therefore start at the top, load speed could be an issue. We then specifically seek out information regarding improving load speed issues on our site. Execute on that information immediately and observe the results. If the loading speed improves but not the conversion rate, then it’s on to the next thing.

We know nothing about color schemes, so we spend some time reading some information about it. After some careful consideration, we settle on a pre selected palette of colors and revamp the site. Observe the results, and move on if necessary.

We don’t know much about copy, so we pick up a copy of Cashvertising and skim the book looking for content that would be helpful. When we’ve found it, we implement it immediately. Observe the results and move on if necessary.

To fix the perceived risk, we can reach out to the 1% that have bought our product and ask for testimonials. So we look up some information on how to craft such an email, and boom send it out. We can also implement a money back guarantee, if we haven’t already.

After implementing all these things, we’re left with a vastly improved site. If the conversion rate doesn’t budge, we merely repeat the process. Come up with additional root causes and systematically test them, looking up information you need along the way.

So to summarize, this is how you rapidly learn new things:
Isolate a problem
Guess the root cause of the problem
Address each root cause in turn
Acquire any new skills to solve that specific problems
Repeat until you’ve got your lambo

This is in contrast to the approach of “oh gee, I have a poorly converting site. Guess I need to read a bunch of stuff on how to improve conversion rates… oh gee, guess I need to know web design, and copy writing, and … oh I’ll just give up”.

This is not to say that reading general business books doesn’t have its place, but I think this is primarily useful for those who are already in motion and can take the 1 or 2 nuggets from a book and apply it to their business straight away. I don’t believe there is large utility in reading for those that are not in motion.

In subsequent posts I’ll talk more about finding underlying principles, and expand on how I’m learning difficult things in my own journey.

 

Dislike ads? Remove them and support the forum: Subscribe to Fastlane Insiders.

[Ninjakid]

When I read the title, thought you were gonna post the Feynman Method, but this is pretty good too

 

[Guest61835]

So I was responding to a thread, and realized that one problem I see with some folks, newbies in particular, is that they lack a systematic methodology for teaching oneself. This isn’t surprising, as it’s not really a skill that we’re taught early in life. Realizing this, I figured I would share what I have learned about learning difficult things, over the years.

Of course, the scripted dogma is that we must go to school for 4 years, rack up a mountain of debt, and hope that we'll have marketable skills at the end. For STEM fields, this actually isn't a bad bet, but since most people don't major in useful things, it turns out disastrously for the majority.

I actually see some parallels between that scripted dogma and some of the posts here. Some folks will do extended learning challenges, work through dozens of tutorials, or seek out mentors to hold their hand... Perhaps it's the case that some people really need these constructs and arrangements, but I believe that there exists a better way to teach oneself.

Full disclosure: I guzzled the scripted dogma for the first thirty years of my life. I loved undergrad so much that I did a 5th year as a victory lap. Not willing to enter the real world, I then went on to a masters program and ultimately a PhD. However, from the ashes of these potentially ruinous decisions came an invaluable skillset that benefits me to this day: the ability to learn new things with minimal guidance.

Much like we have the CENTS framework for evaluating business opportunities, I believe there exists a mental framework for teaching oneself new skills.

What does it mean to learn something, really? I argue that learning means internalizing information in a way that allows you to apply it to new problems that are only marginally similar to previously encountered problems. After all, if you only know how to solve one problem, you’re not really a problem solver, you’re a robot.

Central to learning is having a solid mental model of cause and effect relationships. This is necessary because the world we live in is causal. To illustrate, let’s consider a simple example.

Let’s suppose you’re selling a widget on your website. For the sake of argument, we’ll assume that the widget is something people actually want. You’ve got some traffic, but only 1% of it is converting. You want to increase the conversion rate, because you have big baller dreams of lambos and lavish nights out in the big city. Problem is that with that 1% conversion rate you’re barely making burger flipper money, so hyper cars are out of the question, let alone hiring someone to fix the landing page. It’s on you to solve this problem. How to do it?

Most people will approach this problem by reading blog posts, books, or listening to podcasts on conversion rates. They will think they’re taking action, and assimilating knowledge, but in reality they’re wasting time. There are two reasons for this:

1) This delays the feedback loop between taking action and receiving results, thus obscuring the relationship between action and results.
2) It’s often event focused rather than process focused. Consuming prepackaged information often obscures the why of the solution.

The correct approach to learning how to increase conversion rates is this.

Clearly state the problem; the more precise and detailed the better.
Make a guess about what things you can control contribute to the problem
Formulate a plan for iteratively testing each of these possible root causes. Test most likely cause first, if possible.
If necessary, seek out information that tells you how to address each possible root cause.
Execute the plan.
Repeat until solved.

So if your conversion rate is only 1%, you could guess that the following things contribute to that low rate:
1) the page loads so slowly that people leave before it finishes
2) The color scheme makes peoples’ eyes bleed, so they can’t find the buy button
3) The copy reads like it’s written by an overseas scammer
4) The perceived risk is too high

These are just a subset of the possible reasons, and indeed a combination of them could be the problem.

We’ve already stated the problem, so that’s good.

Our analytics indicate loading speed could be an issue, and indeed any of the other 3 could be the problem so we have no choice but to test them all.

Brute force it is.

We therefore start at the top, load speed could be an issue. We then specifically seek out information regarding improving load speed issues on our site. Execute on that information immediately and observe the results. If the loading speed improves but not the conversion rate, then it’s on to the next thing.

We know nothing about color schemes, so we spend some time reading some information about it. After some careful consideration, we settle on a pre selected palette of colors and revamp the site. Observe the results, and move on if necessary.

We don’t know much about copy, so we pick up a copy of Cashvertising and skim the book looking for content that would be helpful. When we’ve found it, we implement it immediately. Observe the results and move on if necessary.

To fix the perceived risk, we can reach out to the 1% that have bought our product and ask for testimonials. So we look up some information on how to craft such an email, and boom send it out. We can also implement a money back guarantee, if we haven’t already.

After implementing all these things, we’re left with a vastly improved site. If the conversion rate doesn’t budge, we merely repeat the process. Come up with additional root causes and systematically test them, looking up information you need along the way.

So to summarize, this is how you rapidly learn new things:
Isolate a problem
Guess the root cause of the problem
Address each root cause in turn
Acquire any new skills to solve that specific problems
Repeat until you’ve got your lambo

This is in contrast to the approach of “oh gee, I have a poorly converting site. Guess I need to read a bunch of stuff on how to improve conversion rates… oh gee, guess I need to know web design, and copy writing, and … oh I’ll just give up”.

This is not to say that reading general business books doesn’t have its place, but I think this is primarily useful for those who are already in motion and can take the 1 or 2 nuggets from a book and apply it to their business straight away. I don’t believe there is large utility in reading for those that are not in motion.

In subsequent posts I’ll talk more about finding underlying principles, and expand on how I’m learning difficult things in my own journey.

Wow! That's....a crazy observation. So crazy that most people (including myself) were OBLIVIOUS ENTIRELY to their own inability to learn. That's something I've struggled with insanely in alot of things. Except music.
MUSIC is where I found my ability to learn on my own with little research to play instruments. I can play guitar,drums,bass guitar, and such completely by myself. Self taught. Yes I had to research to find the solution and did it but I was unaware of myself being able to self teach myself (or at least unaware of the concept). Thank you!

Now I know where to direct MYSELF into the darkest tunnels with help when needed on certain manners. Thank you SO MUCH for updating me about this post.