Start Small, Start Now: My Journey of Building a "Reverse Advice" App

Start Small, Start Now

I am a backend engineer. Over the past year, I have leveraged AI coding tools (Cursor and Trae) to gain experience in frontend development, giving me the ability to build complete projects independently. I have always wanted to create a product of my own and have brainstormed many ideas, but I always felt the implementation complexity was too high, causing me to lose confidence before I even started.

Later, I realized that these were just excuses. What truly matters is to start acting and solve problems as they arise, rather than setting up imaginary hurdles that prevent you from even gathering the courage to begin. So, I found a method for myself: find an idea with minimal development cost, build it first, launch it, and promote it to users who need it before deciding whether to iterate further.

Based on this, I used ChatGPT to brainstorm a series of simple ideas. Among them, the "Daily Reverse Advice" interested me the most—giving you a piece of advice every day that runs contrary to common sense (e.g., "I suggest you interrupt people more often during meetings today"), purely for entertainment. I thought this was interesting; it differs from the usual advice we hear, and people might smile when they see it.

So, I chose this project as my goal, wanting to experience the process of developing a product from scratch.

One Core Idea

Before starting, I was clear about the product's positioning: to display content contrary to daily advice.

Benefiting from my work experience over the past year, I knew that a product should focus on one core idea, rather than trying to convey multiple concepts simultaneously. Otherwise, when deciding whether to add a feature, you lack a clear judgment criterion. With only one core idea, you can evaluate every new feature: does it conflict with the core idea?

Here is the core idea I wrote down at the time:

We hear too much so-called "advice." Some suggesting what we should do, but devoid of our perspective—invalid advice. In that case, adopting a decision contrary to others' advice can be interesting. This product sprouts from the desire to present "anti-advice" that goes against trends and current narratives. On one hand, it has entertainment value; on the other, it offers inspiration: Why do we follow advice? What happens if we do the opposite?

Focus on One Thing

Entering the development phase, I made a mistake.

At that time, the Gemini CLI had just released a few versions and was free to try. I was curious about the advantages of this CLI-type AI coding tool compared to GUI tools like Cursor and Trae. So, I thought I found an excellent method to merge multiple goals: developing the project while trying out a new product.

After providing my requirements and tech stack to the Gemini CLI, it generated the first version. It did implement development from scratch, generating dependencies and UI components, but it failed to run successfully. The UI page reported errors. I spent time fixing the errors it created, but after several attempts, it still couldn't run normally.

I realized I had to abandon this tool. It wasn't mature enough yet, or perhaps it couldn't satisfy full project generation. I didn't want to invest more time in it; it consumed too much time and returned only an error page and depleted confidence.

That's when I discovered: focusing on one thing is enough. Don't greedily merge multiple tasks, trying to complete them all at once.

I returned to the familiar Trae. After providing the same product and tech stack requirements, the project it built started immediately with a normal UI and no errors. This made me regret the time wasted on Gemini CLI even more.

Subtract, Don't Add

The first UI version generated by Trae had many colors, using gradients to make it look cool. But I didn't want the product to have too many elements distracting user attention.

Product experience requires trade-offs, which also reflect the product's core idea. Since my core idea is to let everyone feel that they "don't need to listen to advice, and sometimes opposite advice brings a more interesting experience," then anything interfering with this core experience should be cut:

❌ Cool gradient backgrounds
❌ Tags for advice types (Life, Fitness, Learning) — users don't need to care about the type; the advice itself is more important.
❌ The total count of current advice

Having a core idea played a critical role at this moment: it helped me subtract from the product.

Design Experience Layers

In this phase, I also learned a method: designing experience layers for the user.

I wanted the user experience to deepen step-by-step, rather than displaying all concepts at once when opening the product. I designed an interaction funnel:

See the Anti-Advice (Simple content, embodying the core point)
View Notes (Supplement the description of this advice)
View Others' Comments (Gain understanding from other perspectives)
Leave a Comment (Participate if you have thoughts or realizations)

Maybe only 20% of people will go from the first layer to the second, and that's okay. It proves this advice isn't of interest to them. I don't want to present the entire experience all at once.

When considering the comment function, I referenced other products:

X (Twitter): clicking the comment button pops up a comment box directly. X is a communication community where speaking and exchanging are primary, so commenting is made convenient.
Reddit: entering the details page to comment. Reddit, I believe, wants to encourage users to see others' viewpoints from multiple angles before participating.

My product hopes to be like Reddit: not guiding users to comment more conveniently, but hoping users can see multi-angle viewpoints to gain inspiration. So, I also chose the mode of entering the details page before commenting.

This discovery surprised me—it was something I could never have predicted before I started.

Meaning Finds You

During the implementation, I questioned the product. Voices appeared in my heart:

"Can't this concept be conveyed just by creating a Reddit group? Is there a need to build a real product?"
"What if no users use it after it's done? Why not do user research first instead of writing code so early?"

These words made my confidence fade, and the initial enthusiasm died down. I paused product development, thinking I needed to find new meaning before I could continue.

After a week, I forced myself to move again. I listed 20 questions about the product and answered them. After answering, I found that meaning doesn't appear from deep contemplation, but finds you when you start acting.

We cannot find the ultimate meaning when we first start. Or even if we find one, it will gradually correct itself during the action, and new branches of meaning will appear. So, don't be hindered by the "sense of meaning"—just do it. Even a small action, paying a little time, meaning will be born in the process.

Fix Before You Create

During the confusion and anxiety, other product ideas would be born. These new ideas made me more enthusiastic, but at the same time, I knew this enthusiasm was temporary.

If I gave up the current product and invested energy in a new one, I would be full of enthusiasm at first, but I would still encounter the current problems. Entering a new idea is very easy, but completing the current idea is more important.

I previously read a viewpoint: Repair is more important than creation. I agree very much. Creating a new one is actually an act of escape—escaping the problem that needs to be fixed right now. But life is long, and you will encounter this problem again. If you choose to escape now, you will still encounter it next time, which actually means zero growth.

Discover Your Weaknesses

I heard a saying before: Investing is something that can discover your weaknesses. When stocks plummet, your anxiety is amplified—will this anxiety change your long-term optimistic judgment? When stocks soar, you worry about missing the next stage of the surge, changing your previous view that valuations were too high. This series of emotional changes comes from investment, allowing you to discover your own weaknesses.

For me, product development is the same.

In this process, I learned about my weaknesses: excessive concern for the future, fear of the unknown, and perfectionism. These weaknesses cause me pain, but they are also what I need to experience.

Looking back at the experience of the past period, it is quite interesting—experiencing the enthusiastic stage, the self-questioning stage, and the stage of harvesting new value.

And these problems that appear can only be solved by oneself. If another person does the same thing, they might complete it faster and better without encountering so many problems. Because our weaknesses are different, the changes we experience are also different.