David Placek
Founder of Lexicon Branding, a brand naming firm
Where to hear and read about David Placek1 episode or test
David Placek founded Lexicon Branding, the naming firm behind Pentium, BlackBerry, Sonos, Azure, Vercel and Windsurf, and his team has named nearly 4,000 brands. The most useful thing to learn from him is to define what winning means, generate far more names than feels needed without judging them, and pick the bold, distinctive one even when the room is split.
There is no power in comfort, not in the marketplace.
Naming expert shares process for creating billion-dollar brands: Vercel, Azure, Windsurf, Sonos
David Placek: the story and the lessons
1David Placek and the Lexicon Branding machine
Placek leads Lexicon Branding, a firm that pioneered brand naming. Lexicon is behind Powerbook, Pentium, BlackBerry, Swiffer and the Impossible Burger. More recent work includes Vercel, Windsurf, CapCut and Azure.
The team has named nearly 4,000 brands, and Lexicon will hit 4,000 projects this year.
He sees a name as the one brand asset that never goes away. Design, messaging and products change, but the name stays. Placek calls the result cumulative advantage, because the bond grows each time people see it.
He uses the Melian Dialogue to make the point: Athens won Melos through preparation, so a company should hold its advantage before the contest starts.
Placek built a linguistic engine behind the creative teams. Lexicon funds research on sound symbolism, including what each of the 26 letters evokes. 253 linguists have worked there over four decades, 108 more work in 76 countries, and databases hold over 18,000 morphemes.
Linguists mostly screen for meanings outsiders would miss. About every third or fourth project, a name is cut and never shown to the client, Australian slang being one cause.
He hires reporters, novelists and speechwriters as namers. Placek looks for curious, hardworking people with low egos, because most of their work gets rejected. Reporters, novelists and speechwriters are used to fast work and rejection.
Agency hires have worked a little less well. With naming software, anyone on staff can reach 200 to 300 names in 3 or 4 days.
2How Placek's team invents a name
Lexicon runs three steps: identify, invent, implement. Identify starts with behavior, meaning how the market acts toward the brand and how the brand acts toward the market, not with mission statements.
Placek says positioning, values and mission worked 25 or 30 years ago but no longer produce names like Sonos. A competitor landscape shows what names already exist, and he calls imitating a form of suicide.
Placek briefs small teams on decoy assignments. Teams of two work under different briefs. One knows the real project, another thinks it is naming for a different company, and a third names something like a bicycle or a car.
Most winning names come from the second or third decoy team. As he puts it, "They are now free to make all kinds of mistakes."
Lexicon generates thousands of ideas before narrowing. Three to four weeks yield 2,000 to 3,000 raw ideas. Trademark and linguistic screens cut them to a short list, and about 250 are real candidates.
Placek dropped large brainstorms because names came from employees and small groups. A short list of 100 names will not clear the legal and linguistic screens.
He makes executives see the name working before they choose. Implementation puts the name on a baseball cap and a T-shirt. A mock Wall Street Journal ad shows a P&G product launch with shares up 10%.
Lexicon also helps write a longer rationale for the three names that go to the president or CMO. About half of projects include customer research.
Placek plans about eight weeks and two rounds of names. Large corporate projects with boards and more presentations take about three months, and approvals plus legal clearance can push them to four.
Clients see two rounds because people learn by comparing, like house hunting where you learn you want a view, not a pool.
3What Placek thinks makes a name great
Placek reads a split room as a sign of a strong name. At Intel, engineers backed ProChip as professional and premium, but Andy Grove chose Pentium because the argument showed energy.
His test is to say a competitor launched as Windsurf and watch the reaction. "Polarization is a sign of strength in the word."
He warns clients they will not recognize the right name on sight. Clients expect to know it when they see it, but it almost never happens. People trust what has worked before, and a name creates an experience for the future.
He points to Harry Potter being rejected 16 or 18 times. Sonos, Azure and BlackBerry were all first rejected by clients.
Placek insists a name must differ from everything in its market. People notice only what is new, like a red pair of shoes in a row of black ones. A descriptive name like Cloud Pro disappears among ten cloud services.
A distinctive name gives an asymmetric advantage before launch.
He aims for an experience, not a word. Google is an experience, while Infoseek is practical and mundane. BlackBerry tells the market they are not like the other guys.
Placek says a word has limited value, but an experience has unlimited value. The best consumer reaction is that the brand is not like the others.
Placek picks letters for the signal each one sends. V is the most alive and vibrant sound in English, B is among the most reliable, Z is noisy, and X is fast and crisp.
Corvette and Vercel use V, BlackBerry gets two Bs, Azure uses Z, and Sonos leans on noisy S. A name needs its sound symbolism and its semantics.
He favors compounds and names the brain processes easily. Compounds like Windsurf and Powerbook combine two sets of imagery, so 1 + 1 = 3. Brains are a little lazy, so complicated names are a liability.
Vercel pairs ver (veritas, verde) with cel (accelerate), which Placek rates perfect fluency. Clients often call compounds too long, so Lexicon researched them and showed imagery to prove they work.
4Placek's naming cases: Sonos, Azure, Windsurf, Vercel
Placek flew back to win over the Sonos founders, unbilled. The founders first refused Sonos because it did not sound like entertainment. Placek argued they make speakers that carry entertainment, and that Sonos is about sound.
It is a palindrome. Founder Bob MacFarlane then said they should name for the marketplace and customers, and later wrote a thank-you note Lexicon still uses in credentials presentations.
He sold Microsoft on Azure by starting a story, not stating a fact. Microsoft wanted a descriptor starting or ending with cloud, and its team first called Azure a dumb idea. Placek warned of an ocean of "cloud this, cloud that" and pitched Azure's sound and its meaning of blue.
His line was "You don't want to make a statement here. You want to start a story." He now puts Azure at roughly a $100 billion brand.
He made Windsurf out of an intangible coding product. The product was called Codium, a name about code. One team only listed words and sports that express flow and dynamic movement, and Windsurf was on that list.
The name was available, and the team took about a week to choose it. Placek says it checks all the boxes, with a wonderful image and a literal physical experience.
Vercel worked because a confident client allowed a coined word. Placek says Guillermo is a V person, and the V fits his strong, opinionated way of working. The parts are simple and familiar, so the name was easy to present.
A confident, innovative team gave permission to coin a new word.
Placek steers AI founders toward tangible names. Engineers lean toward names like Anduril or Anthropic. He spent about $20,000 on research in Europe, South Korea and America and found developers positive and consumers skeptical and worried about their jobs.
That gap led him to advise something people can grab onto, and the Codium team listened.
He points clients back to the market, not their boss. People pick names their boss will like, so Lexicon tells clients, diplomatically, that the choice has little to do with their boss.
It prefers clients who play to win, not just to avoid losing. When the direct client leads and says it can make the name work, managers usually rally.
5Placek's advice for founders naming on their own
Placek starts with the diamond exercise, before any names. Draw a diamond: how do we define winning, what do we have to win, what do we need to win, what do we need to say to win. It can take four or five days.
In a room of five people there are usually five definitions of winning. Startups underrate what they already have to win because they are stressed and busy.
He tells founders to generate 1,000 to 1,500 names without judging. A list of 200 names is not enough. His best advice is 10 or 15 lists of 200 names and directions, then speculate on what each name could do.
Humans survive by spotting what is wrong, and that instinct must be countered. His number one advice is to suspend judgment.
Placek borrows words from unrelated magazines. For a sailboat maker, skip sailing and browse hunting or flying magazines, writing down words and expressions you like. Connecting unrelated ideas surfaces words you would never reach directly.
He bets $5 that two magazines will give you a word that relates to sailing. It is a one-person version of the decoy team.
He tests a name by asking what it could do, not if it is liked. Ask employees what the name could do for the company. Tell friends a new competitor has this name and ask what they think.
Asking whether it fits the concept always produces a descriptive name. Popularity and comfort are the wrong measures.
Placek says the right name comes before the .com. The .com now works like an area code, and AI is making SEO less important. If it is taken, add a small word or use .ai.
With patient negotiation a URL often costs $15,000 to $30,000, and he would rather put that money into the market.
He backs a rename only when the gain justifies the cost. Three situations justify it: a placeholder name picked early for legal documents, a pivot, or a merger.
A larger customer base makes a rename a significant project, so the argument for it has to be made first.
Naming expert shares process for creating billion-dollar brands: Vercel, Azure, Windsurf, Sonos
Lexicon's naming process (identify, invent, implement) and a diamond exercise any founder can run to find a bold, distinctive brand name. It comes from a team that has named nearly 4,000 brands, including Sonos, Azure, Windsurf, Vercel, BlackBerry and Pentium.
Placek leads Lexicon Branding, a firm that pioneered brand naming. Lexicon is behind Powerbook, Pentium, BlackBerry, Swiffer and the Impossible Burger. More recent work includes Vercel, Windsurf, CapCut and Azure.
The team has named nearly 4,000 brands, and Lexicon will hit 4,000 projects this year.
He sees a name as the one brand asset that never goes away. Design, messaging and products change, but the name stays. Placek calls the result cumulative advantage, because the bond grows each time people see it.
He uses the Melian Dialogue to make the point: Athens won Melos through preparation, so a company should hold its advantage before the contest starts.
Placek built a linguistic engine behind the creative teams. Lexicon funds research on sound symbolism, including what each of the 26 letters evokes. 253 linguists have worked there over four decades, 108 more work in 76 countries, and databases hold over 18,000 morphemes.
Linguists mostly screen for meanings outsiders would miss. About every third or fourth project, a name is cut and never shown to the client, Australian slang being one cause.
He hires reporters, novelists and speechwriters as namers. Placek looks for curious, hardworking people with low egos, because most of their work gets rejected. Reporters, novelists and speechwriters are used to fast work and rejection.
Agency hires have worked a little less well. With naming software, anyone on staff can reach 200 to 300 names in 3 or 4 days.
Lexicon runs three steps: identify, invent, implement. Identify starts with behavior, meaning how the market acts toward the brand and how the brand acts toward the market, not with mission statements.
Placek says positioning, values and mission worked 25 or 30 years ago but no longer produce names like Sonos. A competitor landscape shows what names already exist, and he calls imitating a form of suicide.
Placek briefs small teams on decoy assignments. Teams of two work under different briefs. One knows the real project, another thinks it is naming for a different company, and a third names something like a bicycle or a car.
Most winning names come from the second or third decoy team. As he puts it, "They are now free to make all kinds of mistakes."
Lexicon generates thousands of ideas before narrowing. Three to four weeks yield 2,000 to 3,000 raw ideas. Trademark and linguistic screens cut them to a short list, and about 250 are real candidates.
Placek dropped large brainstorms because names came from employees and small groups. A short list of 100 names will not clear the legal and linguistic screens.
He makes executives see the name working before they choose. Implementation puts the name on a baseball cap and a T-shirt. A mock Wall Street Journal ad shows a P&G product launch with shares up 10%.
Lexicon also helps write a longer rationale for the three names that go to the president or CMO. About half of projects include customer research.
Placek plans about eight weeks and two rounds of names. Large corporate projects with boards and more presentations take about three months, and approvals plus legal clearance can push them to four.
Clients see two rounds because people learn by comparing, like house hunting where you learn you want a view, not a pool.
Placek reads a split room as a sign of a strong name. At Intel, engineers backed ProChip as professional and premium, but Andy Grove chose Pentium because the argument showed energy.
His test is to say a competitor launched as Windsurf and watch the reaction. "Polarization is a sign of strength in the word."
He warns clients they will not recognize the right name on sight. Clients expect to know it when they see it, but it almost never happens. People trust what has worked before, and a name creates an experience for the future.
He points to Harry Potter being rejected 16 or 18 times. Sonos, Azure and BlackBerry were all first rejected by clients.
Placek insists a name must differ from everything in its market. People notice only what is new, like a red pair of shoes in a row of black ones. A descriptive name like Cloud Pro disappears among ten cloud services.
A distinctive name gives an asymmetric advantage before launch.
He aims for an experience, not a word. Google is an experience, while Infoseek is practical and mundane. BlackBerry tells the market they are not like the other guys.
Placek says a word has limited value, but an experience has unlimited value. The best consumer reaction is that the brand is not like the others.
Placek picks letters for the signal each one sends. V is the most alive and vibrant sound in English, B is among the most reliable, Z is noisy, and X is fast and crisp.
Corvette and Vercel use V, BlackBerry gets two Bs, Azure uses Z, and Sonos leans on noisy S. A name needs its sound symbolism and its semantics.
He favors compounds and names the brain processes easily. Compounds like Windsurf and Powerbook combine two sets of imagery, so 1 + 1 = 3. Brains are a little lazy, so complicated names are a liability.
Vercel pairs ver (veritas, verde) with cel (accelerate), which Placek rates perfect fluency. Clients often call compounds too long, so Lexicon researched them and showed imagery to prove they work.
Placek flew back to win over the Sonos founders, unbilled. The founders first refused Sonos because it did not sound like entertainment. Placek argued they make speakers that carry entertainment, and that Sonos is about sound.
It is a palindrome. Founder Bob MacFarlane then said they should name for the marketplace and customers, and later wrote a thank-you note Lexicon still uses in credentials presentations.
He sold Microsoft on Azure by starting a story, not stating a fact. Microsoft wanted a descriptor starting or ending with cloud, and its team first called Azure a dumb idea. Placek warned of an ocean of "cloud this, cloud that" and pitched Azure's sound and its meaning of blue.
His line was "You don't want to make a statement here. You want to start a story." He now puts Azure at roughly a $100 billion brand.
He made Windsurf out of an intangible coding product. The product was called Codium, a name about code. One team only listed words and sports that express flow and dynamic movement, and Windsurf was on that list.
The name was available, and the team took about a week to choose it. Placek says it checks all the boxes, with a wonderful image and a literal physical experience.
Vercel worked because a confident client allowed a coined word. Placek says Guillermo is a V person, and the V fits his strong, opinionated way of working. The parts are simple and familiar, so the name was easy to present.
A confident, innovative team gave permission to coin a new word.
Placek steers AI founders toward tangible names. Engineers lean toward names like Anduril or Anthropic. He spent about $20,000 on research in Europe, South Korea and America and found developers positive and consumers skeptical and worried about their jobs.
That gap led him to advise something people can grab onto, and the Codium team listened.
He points clients back to the market, not their boss. People pick names their boss will like, so Lexicon tells clients, diplomatically, that the choice has little to do with their boss.
It prefers clients who play to win, not just to avoid losing. When the direct client leads and says it can make the name work, managers usually rally.
Placek starts with the diamond exercise, before any names. Draw a diamond: how do we define winning, what do we have to win, what do we need to win, what do we need to say to win. It can take four or five days.
In a room of five people there are usually five definitions of winning. Startups underrate what they already have to win because they are stressed and busy.
He tells founders to generate 1,000 to 1,500 names without judging. A list of 200 names is not enough. His best advice is 10 or 15 lists of 200 names and directions, then speculate on what each name could do.
Humans survive by spotting what is wrong, and that instinct must be countered. His number one advice is to suspend judgment.
Placek borrows words from unrelated magazines. For a sailboat maker, skip sailing and browse hunting or flying magazines, writing down words and expressions you like. Connecting unrelated ideas surfaces words you would never reach directly.
He bets $5 that two magazines will give you a word that relates to sailing. It is a one-person version of the decoy team.
He tests a name by asking what it could do, not if it is liked. Ask employees what the name could do for the company. Tell friends a new competitor has this name and ask what they think.
Asking whether it fits the concept always produces a descriptive name. Popularity and comfort are the wrong measures.
Placek says the right name comes before the .com. The .com now works like an area code, and AI is making SEO less important. If it is taken, add a small word or use .ai.
With patient negotiation a URL often costs $15,000 to $30,000, and he would rather put that money into the market.
He backs a rename only when the gain justifies the cost. Three situations justify it: a placeholder name picked early for legal documents, a pivot, or a merger.
A larger customer base makes a rename a significant project, so the argument for it has to be made first.