Makelogo.ai: From idea to $65,000 exit in 3 months
jeannen.com
jeannen.com
Now it’s the era of the AI grift, where every low effort startup which can be made with AI will be made with AI. As long as there’s fast money to be made at the end, everything else is pushed aside. I remember when the norm was “I made this cool thing and had interesting challenges along the way”. Now it’s “I don’t really understand how I made this, but it made me money”.
Case in point, the now infamous 2008 Pepsi logo redesign manual:
https://www.goldennumber.net/wp-content/uploads/pepsi-arnell...
https://www.cbsnews.com/news/pepsis-nonsensical-logo-redesig...
Once you have some validation, sure! But before that, bothering with a designer-tier logo is a waste of time & money. The app is aimed at small indie makers who don't have time/budget for a designer, but don't want a generic vector from flaticons and others :)
I have no qualms with companies starting out and using something generic to make due, as long as they understand the difference. But the dozens of logo maker websites including yours all sell themselves as if they’re the same as a custom logo, and they’re not.
Also, the name and domain do matter. I’ve seen projects fail because they inadvertently named themselves after a negative word in another language.
Some years ago my then-employer saw fit for me to be included in a meeting talking about the company's new logo.
I attended a presentation from a couple guys in marketing, who seemed to be quite proud of what they'd done. It was an hour-long presentation talking about how the relative merits of the subtle hues of blue and orange with oblique angles, versus green and white with rounded edges.
They freely admitted that in addition to spending two or three employee-weeks on this, they had used some...not exactly big bucks, but medium-sized bucks, on a consultant for a week, and felt the results were good and it was time and money well spent.
I then cogently pointed out that it only takes 20 minutes put the first letter of the company's name in the font, select colors and backgrounds, and draw a circle or a polygon or something around it. As proof, I showed my laptop; I'd opened up Inkscape within the first 5 minutes of the meeting and knocked out a perfectly serviceable logo in the next 20 minutes while they were talking about the psychology of the color purple.
I've never made a buying decision for a product in my life based on a logo. And putting this amount of time into the exercise was a foolish waste of time and money; we ought to be building features and documentation.
One of my closest coworkers within that company, who witnessed the whole situation, later explained to me that there is apparently a whole... something to logo design that I utterly and completely just don't get. I don't pretend to understand it, I suppose I felt like a deaf person might feel when someone tries to explain to them what music is.
This was a little bit fascinating, a little bit enlightening, a little bit horrifying, and also terribly confusing.
I didn't have to worry about it much though, because for some reason I was never invited to any future meetings about logos, website CSS, or other branding decisions :)
I'm tempted to ask parent (or anyone reading this) for resources that might help me better understand this world that I was told exists but I know nothing about. I'm actually not too sure if there are tutorials for my level that will help me, or if I'm so "deaf" that trying to learn this stuff is completely hopeless for me.
I called a logo a communication tool, not a selling tool.
But rest assured, every day many many people buy something because of the logo and what it represents. Look no further than the fashion industry: premium brands have their logo plastered on their products in huge sizes or as a pattern because it signals to others a “status”. People buy known counterfeits because they look like expensive originals.
Remember when people badmouthed the iPhone 4S or the first SE because “it looks like a previous year phone” despite being technically superior in every way? Similar logic.
To be clear, I vehemently disagree with those attitudes. But they are real.
To have your logo in a physical product it needs to adhere to a number of technical characteristics. For example, it may need to be identifiable by shape alone, without colour, and not have thin lines because it will be embossed. When a logo is purely digital, some concerns are the same while others shift. For example, you may need it to work at tiny sizes for a favicon, or to display in a watch.
The role of a designer is to understand those necessities and deliver a result which can adequately accommodate them. Just like in software, there are tradeoffs to be made and you need to understand what those are to make something good.
Assuming a 4% safe withdrawal rate, 65k gives you 2600 USD an year. Moving to a country where that is enough to retire on will mean a significant reduction in standard of living.
To give additional context, I worked non-stop for one month to make the app
Other 2 months were pretty chill, I did marketing mainly, which didn't took all my time :)
But before that, I worked non-stop for 6 months or so, I learned to code in May and was learning + making apps to learn non-stop from May to January
And believe me or not, it won't change much
I just prompt Midjourney “in style of logo”
I also pay $8/month but only because I believe I’ll render that many things, which hasnt currently been true
But does this export as vector? That could have value. If not, your app is something my mom could use, while midjourney is not. because she would have to “learn” discord and lose interest far before that.
And yes, the app render vectors :)
Agreed on the "mom" part. Tbh anyone could generate logos using different AI models, but the main point of the app is to offer a simplified experience
I find the writing style a little bit annoying, like reading 3:30am discord message from one of the people I mentor, but I understand it was written as they developed the product? Maybe some editing would have been nice. Maybe using a thesaurus to find some other swearwords to keep it fresh (I'm only half joking).
There is something in there which definitely confirms some beliefs I personally had about people who make such kinds of "startup-y" closed source products. Consider the following quotes from the article:
> I’m trying to convert a curl I found into an Axios function
> some tweaking using chatGTP
> hour spent trying to convert TWO LINES OF PYTHON INTO NODE.JS [...] GTP chat can’t even do it
> spent 1 hour doing research and using GTP to convert the Python code [...] into node
- Commit messages such as "aaazaez" and "aaa"
> I even was able to fully set up AWS S3 using ChatGTP
> sold the app for $65,000
The next few paragraphs may be a little harsh, but I can't help it.
The most surprising part to me was the absolute lack of explanation or even basic excitement about the inner workings of the AI, or any of the technical details in general. To me, this demonstrates a lack of understanding, or even interest, in the core tech of their product. Being able to explain something is generally the number one sign someone understands what they're doing, and this article does nothing but brag and admit to simply feeding everything to ChatGPT.
It confirms my belief that some of these startup-y products, with their shiny UIs, are likely complete and utter horseshit in terms of testability, readability, and likely only work at all because of very lenient tooling and a lick of luck.
There are a lot of things one could have spent time on during the development of such a product, but really, it shouldn't be "converting python to node" for multiple hours, especially when the core AI part took ~5 hours. It's easy to say when you, like me, have never launched a product that made 65k, but it also is not something you (the possibly unqualified reader) should base your workflow off of.
It's very impressive that it worked for them, and that they were able to capitalize on it, and I think 65k can be a load of money depending on where you live. Don't let this cloud your judgement in thinking you can build a high quality app that will actually last more than one major dependency update into the future.
Whoever bought it for 65k either 1) has the money to throw away, or 2) got shafted.
I don't give much details about the AI part because it would have been dumb to make a how-to showing how people can replicate the app
And yes, my code is prob terrible, I learned to code last May, I'm fine with it, as long as the app works
As for the buyer, I'm under NDA so I can't give details, but he has a very good vision for the app and the skills necessary to improve it. It's faster to buy an app already generating revenues and with data than starting one from 0
For how long? That’s the crux of the question. If you don’t know what you’re doing and your code “is prob terrible” the app will break in unexpected ways which are hard to diagnose and fix. If you were a carpenter, would you be proud to sell a chair which held up the two or three times you’ve sat on it but which you can’t confidently say won’t break in six months when your client’s spouse is using it?
If you ever wonder why software today is terrible, that’s a big part of it: people throwing stuff at the wall until it kind of works instead of taking care to make it good and correct which isn’t that much extra effort.
You’re far from the only one doing things that way. And that’s a shame, because we’re all worse off for it.
I was pretty clear when I sold the app that it needed lots of improvements, and that's why I didn't listed it for more (x2.5 total revenues), the buyer know that
And no, I never wondered why software today is terrible, I'm not a dev, and I'll never be, I'm just a guy who learned to code.
But it's good, it means there is a market for people who like "taking care to make it good and correct", and I'll never be able to compete there. Let customers choose :)
This guy is not a software developer and might not be interested in becoming one (he straight jumped into AI). However, he is definitively a skilled seller and despite his lack of a good command of the English language, he was able to compile some garbage with minimal technical knowledge and sell it for $65k.
Welcome to Web 4.0.
23:30 Lol “quick test”.
Well, I found another way to make nice logos. It’s not what I planned at the begining but it will do the trick in the meantime I imagine. Bed, and tomorrow, let’s test the workflow.