If you’ve ever tried to come up with a name for anything, you know what a nightmare it is. When big companies have new stuff to name though, they call in the professionals. Andreas Tzortzis meets some of those who really know what’s in a name...
Thanks to some creative tree trimming, both the Bay Bridge and the Golden Gate are visible from Anthony Shore’s house in the Oakland Hills. You can even pick out the white triangles of sailboats on the Bay and the Blue & Gold ferries crossing from San Francisco to Marin. But Anthony doesn’t. At least not at the moment.
His eyes are locked on the screen of his laptop, where at least 15 browser tabs are currently open. There’s probably more, actually, as every time a word strikes him he opens up a new one to search.
Sometimes it’s Moby’s Thesaurus, sometimes it's Wordnik, a compendium of word lists that word geeks across the internet have assembled. Often, it’s a metasearch engine called OneLook, which plows through hundreds of online dictionaries.
“Mirage,” he says to himself, slowly. “That’s a nice word. I like that.”
Up goes a new tab and mirage gets typed into the Moby Thesaurus. Throughout this process, Anthony will add words he likes to a master document. There are hundreds, likely thousands of these documents on his computer. They’ve been built up over years of naming everything from tech products to investment companies to alcohol brands.
At his fingertips, he has access to databases containing millions of words and their definitions, what they rhyme with, synonyms, compounds, modifiers, suffixes, prefixes and roots—and we haven’t even gotten to the neural network yet.
That this is purely an exercise to showcase his process doesn’t matter. Anthony is in his zone; narrating his pauses, his mouth always moving, his eyes always scanning.
“I’m moving forward because I want to consider these words in their own right,” he says, “and let them take me places.”
Names can’t do everything. Names can’t say everything. The product will always shape perception, as will the visual representation. Anthony utters these caveats before every presentation to clients of his naming agency, Operative Words. And yet names are vital.
Good ones can span product categories (Dove); can anticipate changing business directions (Amazon); can dress up a complex process in child-like whimsy (Google). Bad ones can offend by appropriation (the vending machine company Bodega); leave the consumer confused (the retailer Jet); or alienate existing customers (remember when Netflix tried to launch Qwikster?).
That there are more good than bad ones out there is a testament to the work of a small and in-demand cadre of people who name things for a living.
They come from various backgrounds. Some are linguistics majors, others come from copywriting or journalism. All have an aptitude for non-linear thinking and share a puzzling enthusiasm for plunging down pathways that, more often than not, lead to a dead end.
“You have to have a certain tenacity and excitement for the process, because if you just like the outcome, you’re going to be heartbroken,” Amanda Peterson says. “I think there’s a certain bull-doggedness to everyone who sticks in actual, professional naming. Because 80 percent of the things you work on, never see market.”
Until she “took a year off of naming” last year, Amanda — or Gucky to almost everyone who knows her — was Head of Naming at Google Brand Studio for three years. She was on the team that named Google’s recently created holding company, Alphabet. For a geothermal startup spun off of Google X, she came up with the name Dandelion (“It’s friendly, it’s fun, but the tech also has the structure of a taproot, like a Dandelion”).
A copywriter and brand strategist by trade before she began working for Anthony at ad agency Landor, Amanda comes from the “I want to play bar trivia all day for a living camp” of naming.
“Where you just want to know facts, you know? Like, ok, we’ve got a radio client. I’m going to read all about radio for a week, and I’m going to go through every glossary, and I’m going to learn about Marconi and history of broadcasting.”
Her ability to “speak tech” steered her to Silicon Valley. You might assume tech is the most desirable assignment in the brand naming game these days, but you would be mistaken. Unlike other categories, tech products often do many things, and aren’t nailed down to a specific audience, Amanda explains. A box of laundry detergent or cereal, however, has mere milliseconds to lock you in – and a good name might seal the deal.
“Consumer packaged goods will always be the king of naming: lifestyle brands where people aren’t going to spend a lot of time getting to know you,” she says. “If the product does something nobody else does, you can get away with a name like Facebook, because people are like ‘Oh, okay that’s what a Facebook does.’ It’s not a great name but it fills a need no other company is filling.”
Which isn’t the case for, say, cannabis strains, a naming market that’s arrived from nowhere in the last couple of years and presents some juicy challenges. There are no established trademarks, for example, and you’ll have high-end producers creating names that compete with some common-knowledge street names.
“It has to be reliable, it can’t seem medicinal, but it can’t seem non-medicinal. It has to hit on medicine naming, on lifestyle naming, on reassurance, but also on fun,” she says. “It’s all over the place!”
That, of course, is the perfect assignment for a namer. More than just word nerds, namers love themselves a bit of pop culture, obscure trivia and storytelling. All of it, as Nora Ephron might say, is copy. Amanda, for example, geeks out on the fact that Bluetooth is named after the 1st-century Scandinavian King Harald Bluetooth, who united the Vikings (much in the same way Bluetooth united Intel, Ericsson and Nokia through its short-range wireless connection).
“There are people who are true language geeks and they may get into naming, but they’re going to be limited in the kind of creative they can develop,” says Anthony. “More important is how you think about ideas, and how you see things… It’s about having a fluidity and agility with seeing something from many different perspectives.”
When Anthony Shore was six years old, his parents gave him an American Heritage Dictionary.
“I didn’t have to ask,” he says. “They just knew.”
His fascination with words—their roots, their sounds—was evident. He’d go on to take Latin in the seventh grade and classes in etymology in college.
He eventually transferred to the tranquil redwood groves of UC Santa Cruz in California to finish up a degree in linguistics. While he attended college, he worked as a typesetter and worked on his copywriting at an ad agency. His time working in marketing at a software company in the mid-1990s would help provide the context for his later work.
“You are better at naming if you have a command of broad communications,” he says. “You need to see the potential of a name, you need to envision what its future is going to be. I develop more than 1,000 names for a project, and with every one I’m seeing what that future can be.”
When he arrived at Landor in 1996 for what would be the next 12 years (with a one-year break), his calling and his professional path finally coalesced. He worked on some major projects, including the global assignment to changing Andersen Consulting to Accenture (“The top 53 names went through full legal screening, and each was checked in 65 languages by three native speakers each … It was nuts!”).
But it was only when he left, in 2008, that he felt truly complete. He launched Operative Words soon after.
“There was no bullshit, there was no fluff,” he says, his voice rising. “I was only doing the marrow, you know? My whole life I was pursuing the things I wanted to do and suddenly I was in the thick of only that. Oh my god – it was incredible! It was incredible.”
I’m never going to say the great names are taken because I don’t believe that. I think the obvious names are taken.
Anthony in full flight is something to behold. Bookish by appearance, his presence hums with energy that bursts forth when he hits on a passion point. He speaks quickly and purposefully, traces of his native New Yorker accent still detectable.
You can imagine how he fares in a pitch meeting. Names are personal, as are the associations people have with them. So Anthony makes sure each one he proposes gets its due, photoshopping the names onto the brand website so that clients can see it in context.
“My explanation has a lot to do whether or not they cotton to the name,” he swivels in his chair and adds the word “cotton” to his master Word document. “Because people have their own idiosyncratic associations with names. And people believe that their associations will be reflected and shared by others. Time and again, they say, ‘At first I didn’t like it, but once you talked about it to me, it’s actually really interesting to me.’ It’s important that happens because I need to help them visualize what this name can do for them, and what it can be after this name is launched.”
During an assignment, he’ll disappear into a “word hole” for as long as it takes. Sometimes four hours at a stretch, sometimes 12 – with a couple of days off in between. The deadline is the only thing stopping him from going further.
Ideally, he comes up with at least 120 names for the one thing, preferably 150, which he then checks with a friend of his who runs it through a trademark database. Names that are likely to clear screening go to the client for a first round of consideration. And though he’s developed a couple of strategies for finding names that are free, those times are ending.
“The reality is there are very few words that are available,” he says. “I’m never going to say the great names are taken because I don’t believe that. I think the obvious names are taken. But … the trademark database gets fuller and fuller every day and crowds out names.”
So what is the resourceful namer to do? Enter, the neural network.
Out in Boulder, Colorado, where she works as a research scientist in optics, Janelle Shane indulges in a side hustle. Over the last two years, she’s built and trained a neural network, giving it colorful assignments – like naming metal bands, or craft beers – with often hilarious results. Case in point, the network-suggested recipes she’s cooked (after Janelle fed it hundreds of thousands of recipes) that are – almost universally – inedible.
“It has a limited memory,” she says. “With this limited memory it forgets the title [of what it’s cooking] halfway through the ingredients list. And by the time it’s halfway through the directions it’s completely forgotten all of the ingredients.”
If Janelle’s network sounds like it has the processing power of a toddler, it’s actually much worse than that. It’s most similar to an earthworm, she says, if that worm dedicated all of its capacities to cooking like Alice Waters. What it’s really good at, however, are patterns. And what poses less of a challenge than remembering, is coming up with word combinations that humans might not.
I see a new era of really interesting names that’s just around the corner. That creativity is going to be fueled and led by artificial intelligence.
For example, her list of craft beer names. She added hundreds of thousands of beer names extracted from Beeradvocate.com into the neural network and came up with names virtually indistinguishable from actual names (Who can’t imagine ordering a pint of Flying Rocks IPA, or ducking into the pub for a taste of the Devil’s Chard?).
“The point of naming craft beers is to have a certain rhythm but be a little unexpected, to use familiar words but mix it up a little bit,” she says. “And neural networks are quite good at that.”
When a brewer in Michigan reached out with an assignment to name a new IPA they’d just produced, she added the particular vocabulary the brewer wanted to the neural network and ran it through again. The list of names is hilarious, weird and dark (and definitely worth a look). The winner? A Fine Stranger.
I think we’re going to see new names that are still going to be relatable and curious and familiar and not sound like planets on Star Trek.
Sitting in Oakland, California, Anthony read the post and reached out, hiring her to teach him how to build his own neural network and learn to interact with it. He’s since made it part of his research strategy, letting the network do the heavy lifting, scanning billions of words to come up with combinations that are unique.
Feeding it lists of words associated with science and astronomy yielded cool, yet utterly unique sounding words like “Quarticle”, “Lunate” and “Starly”. That there are still plenty of functionality issues is all part of the growing process.
“I see a new era of really interesting names that’s just around the corner,” he says. “That creativity is going to be fueled and led by artificial intelligence. I think we’re going to see new names that are still going to be relatable and curious and familiar and not sound like planets on Star Trek.”
Of course neural networks will always need someone to guide them. And that’s where Anthony comes in – a non-linear thinker who can disappear down a word hole and leave Bay views wanting; a linguistic who’s crap at Scrabble but great at looking at words from every angle.
“One of my clients once called me a mad word scientist,” he says. “And I wear that with considerable pride, and I have to say it’s kind of true.”