Pick the logo concept that does its job, and save your favourite for the tiebreak. The usual advice runs the other way round. A small business logo has three jobs. It says the one thing that makes you different. It still reads at the size of a browser tab. It stands apart from the marks your buyers compare you with. A concept that fails one of those costs you every time a buyer sees it, however much you liked it in the meeting.
Search the question and you'll be told to print the concepts, pin them on a wall, stand back and trust your gut. A DesignCrowd guide from 2015 on choosing between favourite designs starts well, asking how each one stacks up against the competition. It ends with a coin flip, then with handing someone else the final call. The site's voting poll is there to "gather feedback from your peers, friends, and family".
A poll is fine for choosing a team T-shirt. It's the wrong tool for deciding what your business says to a stranger. Friends vote for the concept that looks best to them. Many of them will never be your buyer.
How do you choose between logo concepts?
To choose between logo concepts, judge each one against the job before you ask which you like. Write down the one thing that makes your business different and check whether the concept says it. Shrink it to 16 pixels and check it still reads. Line it up with your competitors' marks in greyscale and check you can find it. Keep the concepts that pass all three, then let taste break the tie.
Does it say the one thing that makes you different?
This test needs a sentence on paper before the meeting starts. Write what a buyer should believe about you that they can't believe about the next name on their list. Then ask whether a stranger could get near that sentence from each concept's mark and name.
Say you run a bakery and one concept is a wheat stalk. It tells a buyer you sell bread. They knew that from the sign.
If you can't write the sentence, no concept can pass, and the meeting turns into a vote on looks. Stop there and do the positioning work a brand strategist does before you pay for another round of drawings.
If that sentence won't come, the Signal Check is a one-page self-check emailed to you when you sign up.
Does it still read at 16 pixels?
A browser tab shows your icon at 16 by 16 pixels. A social profile crops it into a small circle. Those are the sizes a buyer meets you at, so judge every concept there before you judge it on a presentation slide.
Shrink each concept to 16 pixels and set it beside your other open tabs. Then do it again in one colour, black on white. Thin lines close up and small lettering turns into a grey smudge. What disappears was decoration. What survives is your mark.
If the designer gave you a simplified version for small sizes, test that one too. It's the version on every tab and profile.
Can a buyer tell it apart from your competitors' marks?
Collect the logos of the five to ten businesses a buyer would compare you with. Search for what you sell, where you sell it, and take the names that come up. Put them in a row at the same size, in greyscale, and drop each concept into the line. Look away, then look back. If you have to hunt for yours, so will a buyer scanning a page of results.
Concepts drawn without a brief reach for the same symbols as the rest of the category, which is why contest entries so often look alike. A client of mine, an established retail business moving upmarket, once suggested a monogram built from two letters of its name. I advised against it. The pairing was unbalanced and too close to marks that already existed.
Looking like the others has a legal cost too. The US trademark office calls a likelihood of confusion with another mark "the most common reason for refusing registration".
What if you and your designer disagree?
Build both and judge them against the same job. The same client and I disagreed in the third week of a four-week project. At discovery, the business told me it wanted to compete with luxury retail and to avoid literal symbols of its trade. At the first concept review, the owner's representative liked the type but said the icon didn't represent the field. To me, that asked for the literal symbol discovery had ruled out. At the review of a second concept, the owner said through his representative that I was too protective of the concept. He wanted two letters of the name clearly visible and parts of the mark pulled apart.
My answer was that my job is to advise and explain the risks, and the final decision is always the client's. The money and the logo were his. I also warned that pulling the mark apart would break its geometry and could make it look generic. He was used to seeing lots of options, and I said that showing fifteen random ones is throwing mud at the wall. So we agreed on a test. I would build exactly what he asked for and show it next to the original, and he would decide within 48 hours.
Which version won matters less here than how we agreed to decide. If I were the owner in that room, I'd ask one question of both versions: which one passes the three tests? That puts the argument on the brief, where both sides can check it.
Take this checklist into the review meeting
- Read out the sentence about what makes you different before anyone opens a file. No sentence, no meeting.
- For each concept, ask whether a stranger could reach that sentence from the mark and the name.
- Shrink each concept to 16 pixels, then view it in black on white on a phone.
- Drop each one into a greyscale row of competitors' logos and check you can find it at a glance.
- Cut every concept that fails a test, however much someone in the room loves it.
- If you and the designer disagree, have the alternative built and put both through the same tests.
- Only then ask which of the survivors you like best.
If none of them survives and you're pricing another round, here's what each logo budget actually buys.
