Build a brand from scratch.

A brand is not a logo. A brand is an idea that a logo points at. If you skip this step you'll spend weeks making something that looks fine and means nothing, and you won't understand why it feels empty.

You need three sentences before you draw anything:

• What is it? — "A zine about the detours people end up taking, that also lives as comics, music and merch."
• Who's it for? — "People mid-detour who don't have a map."
• What does it believe? — "The detour is where the good stuff is."

That third one is the important one. It's the thing your logo will have to carry. Vague beliefs make vague logos.

Draw it by hand before you touch a computer. Not because hand-drawing is magic, but because it's fast and you're not precious about it. On paper you'll make twenty bad versions in an hour. On a computer you'll make one, spend three hours nudging it, and get attached to something mediocre.

Rules for this step: use whatever's around, don't erase, fill the page, and don't judge anything yet.

⚠️ Do not skip to Canva. A template gives you something that looks professional and means nothing, and it'll look like eight hundred other brands. Your ugly sketch is worth more than a polished template.

Now you rebuild the sketch as a vector — a logo made of shapes and math instead of pixels, so it stays razor sharp at any size, from a favicon to a billboard. This is what SVG is for, and it's what Code Your Own Badge teaches.

Two things matter here:

The trap is fixing too much. If you clean up every irregularity you'll end up with something sterile that could belong to anybody. Some of the wobble is the character.

This is the longest step and the one every tutorial lies about. You will not get it right on attempt three. You'll produce versions you're excited about, show someone, and realize they're wrong. That's the process working, not failing.

Three habits that make iteration actually productive:

Here's what separates a brand from a logo: real brands need the same mark in many situations. On white, on black, in one ink for embroidery, tiny as a favicon, huge on a hoodie. If you only made one version, you'll bodge the rest and it'll look sloppy.

A minimum system is four files:

• Primary — full color, your default
• One-color — for stamps, embroidery, single-ink printing
• Small-size — simplified; fine detail removed so it survives at 32px
• Dark background — because your primary probably disappears on black

Then lock your colors and fonts in writing. Actual hex codes, actual font names. "Kind of a blue" is not a brand color.

💡 This is the DRY principle from the first tutorial, applied to a whole brand. Define once, reuse everywhere.

At some point you have to stop exploring and choose a primary. Everything else becomes a supporting variant. This is genuinely uncomfortable and most people stall here.

Test your candidates against real conditions, not a nice presentation slide:

• Shrink it to 32×32. Can you still tell what it is?
• Print it in black and white. Does it hold up with no color?
• Put it next to three brands you admire. Does it look intentional or accidental?
• Show someone for two seconds, then ask what they saw.
• Does it still point at your sentence from Step 1?

A logo sitting in a folder isn't a brand. It becomes one when it shows up consistently across real surfaces.

Print shops will ask for vector files with text converted to outlines — meaning the letters are locked in as shapes so the font isn't needed. Free tools like Inkscape do this in one click.

⚠️ Test at real size before ordering a hundred of anything. Print one at home. Things that look great on screen can turn to mud at 2 inches wide.

The strongest brands teach. Not as charity — because teaching does four things at once that advertising can't:

The thing you give away should be your belief in action, not a random freebie. If it doesn't connect to your Step 1 sentence, it's just noise.

Last step. Write it down — not a fifty-page document nobody reads. One page you'd actually revisit. If you can't fill these in, you have thinking left to do, and that's useful to find out now.

💡 Revisit it every few months. Plans that never change were never real plans — they were wishes.

Compressed, so you can see the shape of it:

💡 Count the rejections in that list. That ratio is normal. Nobody's first idea is their best one — they just don't publish the other twelve.

You don't need to buy anything to do all nine steps:

💡 The badge in this guide was built with exactly this stack. No paid software touched it at any point.

Now go make your ugly sketch. That's the actual first step, and it's the one people skip.