Animal Planet's new logo (right) after a recent redesign.

There’s a joke often told in the design world that goes something like this: “The client came back to us and said, that’s great, but can you make the logo bigger?” [Cue knowing laughter.] Those in the design industry know that there’s so much more that determines whether or not a logo is successful. I’m not going to summarize everything you learned in design school here, but let’s just agree that there are a lot of aesthetic factors—color, saturation, contrast, type, type size, layout, complexity, visual hierarchy, etc., etc.,—that dictate the look, feel, and success of an identity system. With much of a logo’s success determined by the specific context of working for an individual client, you might wonder: Is there a shortcut to creating a killer logo? Can anything be learned by taking a scientific lens to the matter?  

A recent study from the Journal of Marketing Research attempted to do just that. Its researchers, Jonathan Luffarelli, Mudra Mukesh, and Ammara Mahmood, “analyzed the effect of logo design”—specifically whether or not a logo is descriptive—on the commercial value 597 companies. The study describes a descriptive logo as one that includes a textual or visual design feature that’s indicative of the product offered; a non-descriptive logo is made of design elements that don’t explicitly show what the company does or offers.

For instance, the National Basketball Association has a descriptive logo: we know by the silhouetted image of a man dribbling a basketball that the logo represents the sport. In comparison, the National Hockey League has logo of a shield, which is much less descriptive. Another example: Starbucks’ previous logo placed the brand name, which includes the word “coffee” around the image of the siren. That made the logo descriptive. When the company rebranded to a pared-down logo that excluded the brand name and removed the word “coffee,” it shifted from a descriptive logo to a non-descriptive logo and asked the consumer to rely on the illustration alone for brand recognition.

The Starbucks logo before (left) and after (right) its 2011 redesign.

The researchers found that descriptive logos can “positively influence brand evaluations, purchase intentions, and brand performance.” In other words, descriptive logos tend to make more money for a company compared to their less descriptive counterparts. The study goes on to explain that’s because descriptive logos are easier to process and tend to communicate “authenticity,” which is a perennial buzzword and something that consumers value. Of course, it’s not that simple.

The study also suggests that companies that offer multiple products or services, like Uber, Procter & Gamble, or the Walt Disney Company, refrain from descriptive logos, as trying to visually capture the unrelated products or services they offer could leave the consumer confused. And if you’re offering a product with a bit less curb appeal (funeral homes or bug repellants, anyone?), you might want to steer clear of a visual reminder there, too. Maybe this explains why 60% of brands today have logos that are considered “non-descriptive.”

On top of all that, the authors qualified that for larger companies, these results matter less because consumers are already familiar with the brand. So what gives? Is there value in high-level, quantitative studies like this?

For seasoned designers, these findings might seem obvious. (One particularly common-sense recommendation from the study: “If you own a coffee shop, you should consider creating a logo that includes a coffee cup with hot steam rising from it.” Sorry, Starbucks). Designers tend to rely on  creative instinct and a deep understanding about how to best expand a client’s audience reach.

The debate of where design sits on the spectrum of art and commerce is not new. But determining where data fits on the spectrum—and how it should be considered—is.

And to be fair, Luffarelli doesn’t scoff at a designer’s expertise, but he also believes when it comes to communication design, the more testing, the better. Testing can uncover quantifiable and clear insights including “confirmation that the design is appropriate and effective, or a problem is identified (in this case a costly mistake can be avoided),” he said. And if you’re curious about what a costly mistake might look like—try a quick search with the keywords “most terrible logos,” suggests Luffarelli. Woe to the logos that appear in that design graveyard. 

Dunkin’s logo before and after its redesign.

For Bobby C. Martin Jr., founding partner of The Original Champions of Design, the preference is always to be more informed than less. And while he’d “of course” consider the conclusions of the study when developing a brand for a new client, he doesn’t take quantitative studies like this one as Gospel.

Martin considers an array of factors when creating an identity system, including individual client research, context, and conversations, and their goals, audience, and competitors. “No two brands are the same, no two people are the same, so no one outcome can define a field,” said Martin. “More importantly, no creativity, no real progress, can happen in the field if we begin relying on formulas and demanding pre-proven outcomes.”

That might leave one to wonder when the conclusions of a study like this are applicable. The annoying answer is: it depends. “It’s always interesting to see how data can inform practice; less interesting to see if/how data can inform creativity,” said Martin. “Design is a balance of the two.”

What’s most compelling about this study isn’t the results, but the questions it leaves unanswered.

That’s the challenge, isn’t it. Design is a studied practice with objective and learned skills, but it’s also a creative art form: It requires subjective and, some might argue, innate creativity; its relative beauty is dependent on individual taste and preferences, on having that certain “je ne sais quoi” that’s not found between the margins of a textbook. The debate of where design sits on the spectrum of art and commerce is not new. But determining where data fits on the spectrum—and how it should be considered—is.

If design is an art, then how do you make a truth statement about something that’s subjective? Likewise, if optimizing a brand’s commercial value is the mark of an identity’s success, then how do market forces influence the value statements we make about design and the kind of product that ends up getting made? Is commercial success design’s ultimate kingpin?  

The big takeaway from the report—that descriptive logos tend to boost net sales—supports the latter theory. But what’s most compelling about this study isn’t the results, but the questions it leaves unanswered: Where does design, and subsequently, decisions about the importance of data fit in? For Martin, design is a combination of both art and practice: “Our contributions as designers is in our specific alchemy of the two.”