How to evaluate a logo

Most logo reviews are not really reviews. Someone presents one large image on a clean background, in full colour, to a room of people who already know the company name. Everyone reacts. Someone says it feels a bit corporate. Someone else likes the colour. A decision gets made, and nobody can explain afterwards what it was based on.

The problem is not that people have opinions. It is that the review was run under conditions the logo will never face again.

A useful logo evaluation looks at how well it works, not just whether it looks appealing in one mockup.

Why the usual review flatters the logo

Take the standard presentation setup and ask what each condition hides.

One big image. At that size thin strokes hold, tight letter spacing reads, and fine detail survives. Shrink it to the size of a browser tab and half of that disappears.

A clean background. No photograph behind it, no busy document, no printed form, no signage with something else nearby. The mark never has to compete for attention.

Full colour. Colour does a lot of work. It separates two words, carries the brand, covers for a weak shape. Take it away and you find out whether the form stands on its own.

People who already know the name. They cannot un-know it. They read the name because they already have it in their heads, so they cannot tell you whether a stranger would.

All four conditions are generous. Real use is not. The logo will appear small, in one colour, next to competitors, in front of people who have never heard of the business.

So test it under those conditions instead.

The six parameters

Each one is a question plus a test. The test matters more than the question, because the test is the part people can disagree about productively.

1. Fit

Does it feel credible for this kind of business and the markets it serves?

How to test it:

Show the logo on its own to five people who do not know the company, and ask what kind of business it belongs to. You are not looking for a precise answer, only for whether the guesses cluster in roughly the right place.
Put it beside the logos of the organisations the business wants to sell to or partner with. A mark that looks out of place next to its intended customers needs explaining every time it appears.
If the business sells across markets, check the colours, symbols and script for meanings nobody intended.

What failure looks like: the guesses scatter, or they cluster confidently on the wrong kind of business.

2. Distinctiveness

Is it recognisable, and easy to tell apart from competitors?

How to test it:

Line-up test. Put the four or five closest competitors' logos on one page at the same size, names covered. Can someone who saw yours once point to it?
Three-second test. Show it for three seconds, take it away, ask what they remember. If the answer is "a blue circle", that is what you own, and so does everyone else.
Sketch test. Ask someone to draw it from memory a day later. What survives is the real mark.
Run a reverse image search and a trademark check before anyone gets attached to an option.

What failure looks like: the description people give back to you fits five other companies in the same market.

3. Legibility

Can people read the name quickly, including at small sizes?

How to test it:

Favicon test. Render it at 16 by 16 pixels, then at 32.
Print it at business card size and at email-signature size, on an ordinary office printer, and read it at arm's length.
Blur test. Apply a small blur, or step two metres back from the screen. Letters that merge at that distance will merge in practice.
Look at it on a phone, outdoors.

Common causes of failure: strokes too thin, letter spacing too tight, lowercase shapes with small enclosed gaps, and names where colour is the only thing separating two words.

4. Versatility

Does it work on the website, in a document header, on a business card, and in one colour?

How to test it:

One-colour test. Black on white, and white on black. If the mark stops making sense, it is a picture rather than a logo.
Greyscale print on plain paper, from the office printer, not from a proofing device.
Place it in every format it will actually live in: the website header, which is usually wide and short; a document or letterhead header; a square social avatar; a signage strip; embroidery or engraving, if that applies.
Put it on a photograph, and on a dark background.
Count the special-case versions it needs. A mark that needs six variants to work will drift within a year, because people use whichever file is closest to hand.

5. Meaning

Do the symbolic elements add a clear idea, or are they just decoration?

How to test it:

Cover test. Hide the symbol and look at the wordmark alone. Is anything actually lost?
Explain test. Ask the designer to say in one sentence what the symbol means. Then ask someone outside the company what it looks like. A wide gap between those two answers means the meaning lives in the presentation, not in the mark.
Watch for stacked metaphors. Two ideas pressed into one shape usually reads as none.

A logo does not have to mean something. It has to not pretend to.

6. Durability

Is it likely to still feel appropriate in several years?

How to test it:

Name the trend. Point at the part of the logo that is clearly of this moment: a gradient, a particular typeface, a style of geometric construction. That is the first thing that will date, and knowing which part it is tells you how much of the mark is exposed.
Five-year test. Picture it on a contract, on an invoice, and on a building, five years from now.
Check it against the business plan. A mark tied to one product line dates as soon as the line widens. A mark tied to a founding location dates when the business moves.

A scoring sheet

Score each option on its own sheet, before comparing. Keep the notes column filled in; the notes are usually more useful later than the numbers.

#	Parameter	What you are checking
1	Fit	Credible for this business and these markets
2	Distinctiveness	Recognisable, separable from competitors
3	Legibility	Name reads fast, including at small sizes
4	Versatility	Works across formats and in one colour
5	Meaning	Symbolic elements carry a clear idea
6	Durability	Still appropriate in several years

The scale:

1 Fails the test outright.
2 Fails under normal conditions, works only in the ideal case.
3 Works, with a caveat you would have to manage.
4 Works in every case you tested.
5 Works, and is an advantage over the alternatives.

One caution: do not average the six and pick the highest total. Legibility and Versatility behave more like pass or fail. A 2 on Legibility is not offset by a 5 on Meaning, because the legibility problem turns up every day and the meaning rarely comes up at all.

How to run the review

Ask for the files, not the mockups. You need the vector original so you can scale it, strip the colour and place it yourself.
Build your own test sheet before the meeting: the small sizes, the greyscale print, the competitor line-up, the dark background. Half an hour of preparation changes the whole conversation.
Have everyone score alone first, then compare. Scoring as a group turns into agreeing with whoever speaks first.
Write down the caveats you are accepting. Every option has some. The ones that cause trouble later are the ones nobody wrote down.

This does not remove judgement, and it is not meant to. Taste decides between two options that both pass. The point is to find out which options pass before taste gets involved, so the decision is not a vote on one picture.

How we work

This is how WizKen approaches brand work. We put marks through the conditions they will actually face before recommending one, and we write down what we found, so the reasoning outlives the meeting and is still there a year later when someone asks why.

If you are choosing between options, or reviewing a designer's work, and want a second read on it, we are happy to look.