The Logo Rabbit Hole

I did not expect the logo to be the thing that got me.

I’m building MyTradePilot, a stock-trading opportunity scanner, using Lovable. I’m 50, with no coding background, so there are plenty of things that should have scared me more than a logo.

And they did.

The scanner button once returned exactly zero candidates. Later I learned the cause was something in a file called mockMarket.ts, involving a broken seeded random-number generator and a hash function returning a float where an integer was needed. I am writing those words now, but please do not mistake that for understanding them.

I just knew the button did nothing useful.

Another time, the app showed a $124,000 phantom gain in the P&L. Not real money, thankfully, but still a horrible little preview of what can go wrong when numbers are wrong. With a trading app, trust is not a nice extra. It is the whole point.

So yes, there were bigger problems.

Still, I kept circling back to the logo.

At the time, the product had already gone through a name change. It started as QuantisAI, which sounded clever in that slightly-too-clever way. Then I changed it to MyTradePilot, which felt more human and more like what I actually wanted the app to be: something that helps spot possible trading opportunities, not some mysterious black box pretending to know the future.

But the name change created its own mess. The homepage said MyTradePilot™. A learning page said TradePilot Academy. Other bits still had old branding floating around. It felt like walking through a house where every room had a different family name on the mailbox.

Naturally, instead of calmly fixing all of that, I decided the logo needed work.

I paid for logo generation. Then I rejected the first one.

Then I asked for five more alternatives.

Then I studied them like I was choosing a flag for a new country.

One concept had a paper airplane mixed with a candlestick chart. For anyone not living in trading screens, candlesticks are those little red and green bars that show how a stock moved during a time period. The paper airplane idea appealed to me because of the word “Pilot.” It felt light, directional, simple. Not too Wall Street. Not too robot.

I picked it.

Then I changed my mind.

This is the part where I have to be honest: I probably spent too long on this. Actually, not probably. I did.

There I was, with real problems still sitting in the app. The scanner had to produce useful candidates. The P&L math had to be correct. Security scans were throwing phrases at me like SUPA_anon_security_definer_function_executable and search_path_mutable, which might as well have been ancient runes. I had to trust the AI to fix things I could not independently explain.

And meanwhile I was fussing over whether the little airplane felt “too playful.”

Eventually, after all the paid generations and rejections and second-guessing, I uploaded my own image.

It was not a grand design strategy. It was more like reaching the point where I needed the loop to stop. I needed the app to have a face, even if it wasn’t perfect, so I could get back to the parts that actually make or break trust.

I think the logo became a safe place to worry.

When the issue is a security warning, I feel out of my depth. When the issue is a phantom gain, I feel real fear. When the issue is market data and recurring monthly costs, it feels like I’ve crossed from tinkering into commitment.

But a logo? I can have an opinion about that. I can reject it. I can ask for five more. I can say, “No, the wing is wrong,” and feel briefly in control.

That is the trap.

Design matters. Names matter. First impressions matter. I still believe that. But at some point, polishing the front door is just a way of avoiding the wiring in the walls.

If I were starting again, I would pick the name earlier, clean up the branding across the site, choose a simple logo, and move on before my brain made it into a referendum on the whole product.

Small takeaway: sometimes the tiny design decision is really fear wearing a nicer shirt.