Do Field Tests

by Blake W. Howard

In the world of verbals & visuals, subjectivity reigns. So how, in a discipline filled with varying opinions, do you know for sure something will work? I call them field tests.

I often say people don’t pay our firm to design they pay us to help large organizations make decisions (and hopeful wise ones). A great way to help groups avoid camels (a horse designed by a committee) is to show hard driven data that proves a proposed design will work. Field tests can provide that data.

Warning: Proceed with caution. Testing on it’s on should never dictate the final decision for a mark. It is simply a data point to consider. Traditional focus groups are great for gaining valuable insight into consumer behaviors, but terrible environments for directly testing logos, names, or even taglines. People will always want more of the same. Henry Ford said, “If I’d asked people what they wanted, they would have asked for faster horses.” They only know what they know.

Simply showing someone a logo and saying “What do you think?” is an injustice to the work and a fast way to see beautiful work die a slow death. Logos and the like never exist in a vacuum. They will always exist in a some sort of context to be considered (a logo on a business card being handed to you in the hand of man wearing a cheap black suit standing in line at Starbucks with a cheesy grin on his face and latte foam in his mustache, for example). So to test names, taglines, or logos out of context for opinionated likes and dislikes is an inaccurate measure for effectiveness.

Below are three ways, I’ve found, that you can properly test ideas.

3 Helpful Field Tests:

  • Relevancy: Does your target audience understand the core concept in which you are trying to communicate? For example, if you are testing the name “Jam”, ask people similar to your target audience what they think of when they hear the word or phrase. “Jam” could be on a sandwich, it could be getting into trouble, or it could be a new hot “Jam” on the radio. If they answer closely to your intended concept then it is a positive response. This is best done in a one on one setting, so group think doesn’t taint natural responses.
  • Memorability: Discovering if a mark or a name is truly memorable is monumentally important for success. You can find fun ways to trick a test group into thinking they are being tested for one thing when in fact you are testing them on something else. For example, if you are testing a logo aimed at 5th grade students, have their teacher show them ten logos in the morning and talk about their favorite. At the end of the day, the teacher can have the students write down what they remember. If some of your top candidates made the cut then you know you’re on to something.
  • Distinctiveness: Knowing if a brand mark or name is truly perceived as different from the competition is a great way to ensure success. One way you can do this for a logo is to mock-up a proposed design in a sea of competitor’s logos. Usually one can tell quickly if it is distinct, but if more data is needed simply poll near by observers for a helpful set of data. Another quick test is to through your new proposed logo onto a competitors ad. If it feels awkward and disconnected then thats a good thing. If it feels interchangeable that’s bad.

Testing subjective material is always hard, because it is not black & white, it’s quite grey. No two tests should be the same. They should be contingent on the material and audience you are testing. The goal is to dig beyond opinions and boil it down to test the timeless and proven principles. There you will find gold.

Make marks that are distinctive. Place your work in a matrix of competing imagery to see if it feels distinct. You can also try placing a new mark in a competitors ad, if it feels out of place, you’re on the right track.

Illustration courtesy of Craig Winzer

Dig Deeper!
Graham Smith, logo designer from imjustcreative, puts the Logo Test in action with his Brand Reversion series:

More on Logo Tests:

Some interesting “Heat Mapping Tests” with Logos: