The Art of Figuring It Out: How I Solve Problems Before I Ask Anyone

This is a personal post about a habit that changed how I work. It is not a technique for any one tool. It is a way of dealing with every new task that lands on my desk.

The question I ask first#

When something new arrives, such as a video to edit, a graphic to design or a system I have never built, I ask myself one question before I ask anyone else: have I already tried to find the answer myself?

Have I searched the internet? Have I looked at what exists? Have I asked an AI tool? Have I used my own head to work out how I would do it?

Only after all of that, if I still have no answer, do I go to another person.

Why I do it this way#

Waiting for someone to give me direction does not work for long. "Tell me how it is done," "teach me," "give me some direction": asking that way means my progress depends on someone else's time.

And here is the thing: everyone has the same tools. The same opportunities, almost the same subscriptions, and the same free, open internet. Anyone can use it, and anyone can use it well. So the difference is not who has access. It is who bothers to figure things out.

My process, step by step#

1. Check yourself first#

Before asking anyone, I check: did I search? Did I research? Did I look at it properly? If the honest answer is no, I do that first.

2. Use the internet, and use several sources#

I do not trust one source. I look for three or four before I rely on something, because one source is a claim and several sources agreeing is evidence.

3. Use AI, but never just one tool#

I am wary of how people use AI. Asking one tool one question and taking the answer is not research. AI makes mistakes, and those mistakes sound confident.

So I do not rely on one. I try five or six different AI tools, take all their outputs, and then make a single judgement. That is a council of AI: several opinions that check each other, with me deciding. (I wrote about a skill that does exactly this if you want the structured version.)

The point is not that the council is always right. It is that a mistake in one answer tends to stand out against the others, and you can catch it.

4. Summarise it#

After the research, I boil it down to the essence. What did I learn, what do the sources agree on, what is my conclusion? Once I have that, the research is done and I can start using it.

5. Make something#

Now I make a first version: the video, the design, the plan, the system. It does not have to be perfect. It has to exist.

6. Show people, then refine#

Only now do I go to others. I do not say "tell me what to do." I say "here is what I made, and here is how I got here. What is your view?"

Their input makes it better. Then I refine it, and show it again. Then again. This loop of make, show, refine is the part that gets the real quality.

It has worked for me every time so far.

Why being a jack of all trades works for me#

There is a saying that a jack of all trades is no match for a master of one. I am not embarrassed to call myself a jack of all trades, because I think it is more useful than it sounds.

What I understand is the whole pipeline: from design, to video, to implementation, to marketing. I know how email outreach works, how design works, and what actually moves people. When you understand how each stage affects the next, you stop being a specialist at one step and become someone who knows how things get done.

I think that is a powerful place to be, because you can see where the impact comes from.

Take a single post. A well-made, striking post can do far more than a generic one that an AI produced in one click, and that is no accident. We have all seen Instagram pages that run bold, creative ads and gather followers because of it. The difference is rarely the tool. It is the person who understood what makes people stop and look.

The loop, in short#

1. Check yourself: did I search and research?
2. Read widely: three or four sources.
3. Use a council: five or six AI tools, then one judgement.
4. Summarise: reach a conclusion.
5. Make a first version.
6. Show people, take their view, and refine, again and again.

A caution#

Figuring it out first is not about doing everything alone, and it is not an excuse to stay stuck for days. Set a time limit. When it is up, take what you have learned and tried to someone else. You will arrive prepared, and they will be able to help you quickly.

The point#

I am not waiting on anyone. I work on myself and on the output, and it keeps getting better. Everyone has the same tools and the same chance. What matters is how much you are willing to consume, figure out and make before you ask.

That is the framework I use, and I think everyone could.