A practical guide to Lifelong Self-Directed Learning


Yesterday night I shared the dinner table with 3 young Japanese at the backpacker’s lodge of Alice Springs. They just came back from Uluru and were on their way back to Japan. One faced me and asked. Could you share your trick for work? When I’ll be back in Japan, I start right away my first work at NEC (A large Japanese Computer firm). I’m sure you can share a valuable trick I could use.

At first, I claimed that I was not qualified to tell someone about work. My job is to create software not to teach others. I also told him I was a lifelong entrepreneur, I have rarely been in the position of an employee.

Nevertheless, the discussion continued over an hour. We talked about sales, being organized. I told him about the rollercoaster life of an entrepreneur and that it all starts with having an idea to sell. I compared with the daily life of a traveler, something we both shared. I shared a list of books and persons to follow in marketing and business development. We ended up with a “see you again” although we never shared our names or contacts.

The following morning, before leaving, he came back to tell me: You see, you were wrong. I was sure you had something important to share and I was not disappointed.

Do I suffer from Impostor Syndrome? I’ve been sharing my knowledge with tons of people. Knowledge of Life, work, and learning. Nobody complained it didn’t work. I lack theoretical background but I have a lot of real practice. I learned a lot. I keep learning to do my job. So what’s my secret? A good lifelong self-directed learning process. Not a fancy trick to hoard more but a practical tool to learn to do more.

Learning as you build cathedrals

You’ll find plenty of literature about designing your learning plan, drawing a line that goes straight from where you stand to the ideal skillset for your dream job. This mimics Cathedral Construction and dreams of School time formal learning. Let’s set goals, build a plan and execute. I’ve never learned this way since I left University and rarely seen others do it.

There are two obstacles and several contentious points


It’s hard to stay motivated for such a long time without tangible results that prove we are on the right track. How do you know you are on the right track? When you start being able to execute tasks at a decent level of performance and quality. Being able to follow conversations of reputed people in your art is also a good indicator that you covered all the topics.

When you build a cathedral or write a novel you have to stay motivated until the last moment. Lifelong Self-Directed Learning doesn’t need to be that hard.

To make it less hard we need intermediate tangible results. Agile Methodology taught us about short iterative loops. Why not reusing it?


To be self-directed you need to understand directions. Imagine you are in the middle of the desert and you want to reach a point. This happens every day for me this week. You take out your smartphone with a map you downloaded before (No cell towers in the desert FYI ). Now you have both where you are, which direction you face and where is your destination. So well equipped self-directing yourself is easy.

Compare with learning.

You have what others share they did. If you had a map, it would be like hundred of paths from start to end but none are fully documented, trustable or even realist. Most people who share overnight success plans are boasting. Real learning stories are boring to read because it takes a long time.

You don’t know where you stand.

You could test yourself but it’s just testing material you just learned. The best is to move around your position and see how the world react. This means applying what you know and see if you can do it well. In a way, you are the GPS and the compass needle. It’s your interactions with the job and others that let you know where you stand. If you don’t move or interact, you don’t know where you are. This happens too when on your GPS the North is not calibrated. Happens when you change hemispheres.

You don’t know how to get there.

A learning plan is not flat, it resides in an abstract space of high dimensions. Distances don’t follow Phytagore Theorem. The shortest path is not always the straight line. Detours may happen. There are black holes. The topography may change as you learn and science evolves, quickly. The only self-directed solution is to try, carefully and learn from experience.

So-called learning experts will tell you they know how to design a learning plan for you. Some will also tell you most learners can’t direct themselves. This is handy: it keeps them in the loop and with work. The truth is that it’s hard and must be constantly tested. It requires attention like driving on a mountain road.

I guess that’s one of the benefits of having a mentor. You have someone looking at your progress from the endpoint. Being mentored is not being self-directed. I digress.

You don’t even know what is there

By definition, you don’t know what your destination looks like since you are learning about it.

You can mitigate that. Gather some ideas from the web. Try to find a few people who reached this mastery and follow them on Twitter or on their blog or on Medium. Twitter lets you lurk them at length. On Twitter you can engage directly with them without being a friend or introduced. By observing their language, what they talk about you will have good insights on what it looks like to be an “your dreamed job here”.

Your short list

It’s good though to make a little plan when there is a lot to learn.

At this stage, building a complete list of skills and a plan to learn them all is irrealistic. What you need is a progressive process that let you gain a clear image of your destination while starting to learn. Take some of the names, concepts, and techniques mentioned in recent blog posts and find a few pages to read about it. That will be your initial learning plan. It will bring you to a better understanding of what you don’t know and what it is about.

It must be a few lines.

Here is what I shared with my Japanese friend who wished to become an entrepreneur

– Personal MBA ($15) to get the top concepts of how to operate a business
– Lean Startup to get an idea of how to start a business today
– Business Model Canvas to get an idea of what is needed to build a business plan
– Follow Seth Godin Blog (and his alternate MBA) to get an idea of what’s coming next
– Follow Rebekah Radice for quick applicable tips and Social Media selling practice

My advice was to loop through those resources and try to find occasions to use them and/or apply them to understand his new employer strategy.
– Read
– Understand
– Apply to real cases he encounters
– Expand knowledge
– Iterate on a new opportunity of learning (self-feedback, network, curiosity), keep a log of what is not explored

It looks a bit unstructured but it’s lean. He will avoid learning things he won’t apply. It’s constantly validated.

Anyway, learning messy. You don’t control what you will retain and how deep you will understand at first exposure.

The Virtuous Cycle

I’ll share a practice that worked well for me, designed after Software development good practices from last 10 years. We don’t build cathedrals anymore, with long waterfall project plans, milestones, partial delivery, betas. That’s old time, last century business.

The idea is to make small iterations in form of circles. Each iteration brings you in a better position to start the next one.

You are
– More motivated
– More aware of where you stand
– Know what worked or not
– Made tangible progress in the skill
– The job has been done while you learn, it’s sustainable forever.

In an Agile Scrum or Cycle, there are multiple phases. Let’s explore what could be the phases of a Self-Directed Learning agile cycle.

I see this as the equivalent of TDD (Test Driven Development) and CD (Continuous Deployment) for SDL (Self-Directed Learning).
Let’s say CI stands for “continuous improvement” (integration in fact), similar to living in beta but for the pros. It’s a virtuous circle with small agile loops going from planning to deployment (CD) in short cycles. That’s the modern view of development.

In TDD you write the tests, then only you build the program that satisfies the tests. Tests are supposed to be easy to write and we are still objective on the desired outcome.

I’ll learn and repair this sink. Water will no more leak, there will be no visible marks of my repair.

The agile alternative consists in writing stories describing what will work after a cycle.

We could translate it into statements as follow for self-directed learning.

After learning I will be able to cook this recipe in 1 hour and obtain a dish with a recognizable taste although texture and aspect may not be perfect.

After re-reading the book Business Model canvas I will redo my business plan and have something I dare showing.

Sometimes the learning phase can be only in the intensity of focus, dedication or effort during the doing phase. A constant learning from experience.

I’ll stay super careful while writing this piece of code. I will keep a journal of what I do. I’ll prepare and keep next to me a checklist of what to control. At the end, I will have a nice result as if I master this knowledge. The performance will not be great but at least I’ve done it once according to state of art.

the next iteration could be

I’ll redo a similar work without support from documents and review my work from a learning perspective. How much I learned, what is left to memorize.

this next iteration could take place sometime after. that’s where the journal comes handy, to reload context and knowledge faster.

For the pros, this looks a bit like a learning objective but without the formalism, anchored in real life and with measurable aspects made prominent. The measurable part takes place after doing the job. We don’t judge on the potentiality of doing but on the actual job done.


We will learn the recipe of “Coq au Vin”, see a demonstration and verify with an assessment we got it right.
We will learn the recipe of “Coq au Vin”, then prepare one each by ourselves, taste it and infer if we learn it right


It’s very rare that you can start learning without a formal step. Learning from books, blogs and videos is a good way to enter the subject quickly. it should remain short. The see will be a formal piece. The virtuous circle should start as soon as possible after this lead-in.

All the phases

I came with the name of “Application Guided Learning”. It’s the full circle of learning by doing:

1. Set a goal of something you want to be able to do.
2. learn something about it
3. do and check you learned by doing
3. understand what is needed from failures, slow performance, and poor results

1. Plan improvement in result or execution
2. search for more knowledge sources, learn more
3. reapply or enhance last work
4. see where you stand and look at the map

when you reach a decent result, proceed to next task and start again.

PS: It could be next task is already well know. Learning cycles are embedded into real work and fit seamlessly in our work plan.


You don’t plan in the abstract based on what somebody told you, or a curriculum you copied. You have a general direction given by the people you see doing the work. You wish to practice as they do. This leads to a work plan you estimate (a work plan, not a study plan). We focus on the next task. You set your new learning goals from what you can reckon is missing in your understanding of the concepts or in the execution of a task. This means each learning goal is very near and connected to where you stand. No big jumps, no plans on the comet.


Job to learn: prepare an omelet.
Next task: break eggs.
Learning goal: How to break eggs safely without adding shell bits and verify they are fresh.
Break eggs as learned, this implies repeating 3 times
Did we perform well? If no reread instructions. If we broke them all put back in my log to practice again.
Proceed to next task: Mix eggs

How to identify what you need to learn?

You see people and documents mentioning a concept you totally ignore? Dedicate time to sort out what it is.

Your job or your prototype of job requires a skill, or you did poorly for last execution of a similar task? upgrade your skills by learning formally or from others before trying it again

Your last execution was slow and imperfect? Review your knowledge and prepare to try again with more intensity

You did sufficiently well? Proceed to next step of the work, set new learning goals for this step. often it will be a slow step by step execution with your eyes fixed on the book.

Search information

It’s really searching and validating the best source. We rarely learn from 4 sources at the same time. It would require an effort of synthesis we can hardly do as we are in the process of learning. So we search and bet on the best source we find. It doesn’t matter if it ends up not being the best because this is just one iteration. Often a subsequent iteration will use another source which looked too arduous for the first iteration.

I often keep the pointers to other sources in my journal and the reason of my choice. It’s important when I review my learning at the end of the cycle. A failure could be due to a bad choice of source.

Learn more

There will be prereading and learning before doing. It can be as short as rememorizing during the search. It could be sitting in front of a video for 20 minutes (not more) or reading a chapter. I rarely read more than an hour before applying.

This learning is like applying a violent flash of light to the job ahead. It will shed light on what we have to do in order to reach our goal with acceptable quality and decent execution.

More learning can happen during execution. It will be more in the form of job aids, short reads, gleaning more information from Q&A sites or step by step execution.


Imagine the scene. You have the recipe book in the kitchen. It’s maintained open at the right page. A sheet of plastic protects it from the splashings. You are hesitant. You wear a long apron. You got the feeling. This is learning by doing.

The same feeling is present when you do your first business plan, your first react native program, your first plumbing repair, your first web design.

It’s a combination of hesitant execution, with a lot of attention on the guide and a lot of intensity.

This is where the Application driven takes it meaning. You learn in view of executing properly.

– Keep an eye on the result. Estimate your chances that the result might conform. If you broke 2 eggs on 3 and the result is full of shell bits you are heading for disaster. Stop, relearn, find additional sources with more details or explanations at your level. On need start over with a new bowl.
– It is work to be done. it is your job but you are a learner. Some provisions must exist in time to allow failures. Beware deadlines, busy schedules that would force you to end at all costs and not let you learn from your experience.
– Going back to the center the road request extra attention and effort. Expect Learning while doing to be more focused and tiring than watching a video on your couch or doing your regular work. You must apply force to the wheel to stay on track.

Learn from Doing

Learning is something you do to yourself. It changes your state of consciousness. You can’t really force learning to happen but you can create the best conditions for it to happen and you multiply occasions.

– Observe yourself doing. Keep a journal, speak out loud.
– Take breaks, suspend execution to think about what is going on.
– Congratulate yourself for every success. Associating pleasure and proud with learning helps the subtle alchemy of learning to take place.
– Share your work. Others may ask questions that increase your level of consciousness. The sooner the feedback, the deeper the learning.

Learn from your learning experience

The task is finished.

Two important questions:

– Did I do it well? Our goal here is to judge our level of conformance to our learning goals. This will be help us know if we reached enough learning on this topic.
– What is efficient? Did I overkill it by learning too much in details, doing too slowly, learning too many related aspects. Did I failed to choose good sources? This will help us perform better at the next iteration.

and the question on the money:

– What should I learn next?

Since we proceeded in such small steps, it’s pretty easy to answer.

– Look back on your experience and see if you experienced knowledge gaps. Parts of the job you did without understanding, following instructions almost like a robot. I have this very often in coding. I copy an example, read the explanation but don’t get it. It’s a distraction on what I’m currently learning so I apply as is. I need to get back to it one day. I will add this to my learning backlog. In coding, we have a name for this: Technical Debt. Shall we say that a knowledge gap acknowledged in the middle of something you want to learn is a Learning Debt?

– A sidetrack you may want to explore. You didn’t need it but it could help expand your mastery by owning the sides as well. Add to Backlog too. It’s not a debt since you didn’t use it.

– Something you totally missed. It was right on your path, you bang on it, it worked but you don’t know why. It’s a debt but dangerous since you don’t know how to repay it (aka learn). Some urgent action should be to apply some research to uncover the extent of your lack of awareness. Learning could take place after.

– From where you stand, analyze the next task and see if you foresee things to learn before starting it. It’s often that it’s only at the point of need and with the experience of the preceding tasks that we can measure how much we need to learn.

Take corrective action

Spaced repetitions

No learning is final without 3 spaced repetitions. Repetitions must include some retrieval practice.

My advice is to keep track, in written or with sketches. Using your fingers to keep track will help retention. It has to be easy and fast to skim and watch again. Later when you will sort, reorganize your notes, your brain will practice retrieval and reinforce retention. Videos won’t do it.

Review your notes and tracks when you restart the next task or if ever redo the task. This will let you reconnect with details and aggregate experience you gained in between. Take notes again.


The virtuous circle and process I describe is only a template. This means that you should fill it with the actual tasks that will be the doing part. Those tasks will impact the duration and frequency of the cycle. The job to which you apply this system will change the loop.

If you learn cooking, each application of knowledge will rarely extend beyond a few hours. You can’t repeat the same recipe again and again until it’s perfect unless you are studying in a chef school. It would be boring, you can only eat two meals per day and your taste would be blurred quickly. This means you have to interleave multiple learnings in parallel and iterate on the same practice on a cycle much longer than execution of job requires. The cycle will be paused for a few days.

If you learn to code, each application of could take days. Your cycle will be timed in days or weeks with learning taking easily 20% of it.

If you learn to write, each application will in hours, your cycle can be as short as a day and there are not obstacles to repeat it immediately.

If you learn Yoga, each application will take an hour, your cycle will stop and restart the next day.

So stay flexible. What counts is the idea of iterations made of phases of learning, doing, validate learning, planning new learning based on acknowledge needs, with a general direction in mind and observing others.

This template will be applied a lot of times. Multiple learning can take place in different places during the same period (but rarely simultaneously). If you have to take a picture it would be like plenty of concentrical arrows, each rotating at their own speed, pausing for whiles. There is not one virtuous circle or a spiral of learning patiently taking place once after the other. There are 10 or 15 circles possibly forking, converging and merging when topics reach complex domains or synthesis takes place.

Don’t start too many and end with cognitive overload. Each must reach their point of application. They can only be suspended between tasks. Using a backlog ( a long todo list) to keep track of what was paused or not started.

Final Thoughts

Strange I didn’t tell my Japanese friend to keep learning, always learn. Well, we talked about Ikigai, our same-day #PKMChat topic, which implies it. No need to repeat was is already known.

Will you follow my advice? Give it a try and keep me posted.


Rollercoaster life of the entrepreneur via rebekharadice