Naming Strategy
A name is a word that uniquely identifies a specific person, thing, or group of things (like companies or product lines). But almost all names do more than just identify: they describe. When coming up with a new name, your most important decision is to determine what you want to say, and how you want to say it.
The industry standard way of answering this question, unfortunately, doesn’t work. I used it for years before my freelance days, and I’ve seen things you wouldn’t believe. I’ve seen briefing processes leave clients unclear on what they’re going to see next, making it impossible for them to review or approve work confidently. I’ve seen creatives interpret exactly the same brief in opposite ways, and neither of them the one that was intended. I’ve seen the creative process draw out over round after painful round, with hard pivots as the team struggles to figure out why their names aren’t working, and what they should do instead. I’ve seen a client who thinks they’ve finally seen it through have their hopes dashed as they struggle to communicate with their legal team and get their name trademarked.
Fortunately, It doesn’t have to be that way.
I’ll talk about a system that works, why the industry standard doesn’t, and how to handle this as naming pros.
A system that works: Message and Method
All names share a pair of goals: to convey information about the thing it names, and to stand apart from other things like it. These goals are often in tension, and balancing them is the art of naming strategy.
Rather than target these goals directly, namers set criteria around the name's message (what does it say?) and method (how does it say it?)
The system, part one: Message
A name's message can be almost anything that is important about the company or product, as defined by your brand strategy. For our purposes, these ideas almost always falls into one of five buckets:
Message 1: Who
Names that tell you who made this thing, either using a human name, like McKinsey, Chase, Chanel, or Ford; or a place name, like Texas Instruments or Maple Leaf. These names often evolve into acronyms, like P&G, BBDO, EY, or TD. This is also where I'd categorize names intended to be meaningless, such as Kodak or Noom: if nothing else, it tells you what company made it. (By that reasoning, this is arguably a de facto sub-message of every company name).
This was once the default main message for company brands, accounting for about half of names up to 1930. It's very rare today, accounting for less than 5% of new company names in the last ten years*.
Message 2: What
Names that tell you what the thing is. This obviously includes a lot of real-word noun-based names, like Home Depot, The Container Store, Apple Watch, Google Calendar, or Amazon Basics. But these names can take on a range of forms: at the other end of the spectrum are numeral-based systems used by car companies like BMW or Tesla, where the BMW M3 has similarities with other Ms and 3s. (Personal opinion: if you have to publish an article called How to decipher the BMW naming system, you've made a grave error. BMW, please contact me, I can help.) In between are verb-based names like Pampers, that describe either what the offering does or what audiences can do with it.
This is the default approach for product naming, and is really the only message that many branding agencies consider, as we’ll learn about later.
Message 3: How
Names that tell you what the thing is like. This is sometimes conveyed literally, as with Seven-Eleven naming themselves after their store hours, or in names like The Honest Company, EasyJet, Uber, Total, or Progressive. Many less literal names, like Microsoft, Sprite, Anthem, or Amazon target this message.
This is a relatively common message, although it's more often conveyed evocatively or cryptically than intuitively (more on those terms below).
Message 4: For who
Names that tell you who the thing is for. This can be conveyed literally in a name like Salesforce or Peloton or Adobe Illustrator, or evoked through names like Venus.
It's a very uncommon, cutting-edge approach with no strongly established use cases, though it makes the most sense for entities targeting a unique audience niche or permutation.
Message 5: Why
Names that encapsulate a benefit, telling you what you get out of it. This might refer to a desired action like Excel, an end state like DraftKings or Venus, or a feeling like Yahoo! or Yum Brands.
Like audience-based names, this is an uncommon approach without a strongly established use case, though it makes the most sense for entities with a clear universal benefit.
Again, in developing your brand strategy, you'll identify the credible, relevant, differentiating features to build associations with, and the most important of those will form the core of your name's objective. Your name will convey the top few, maybe from the same bucket (like The Container Store, all about What), or mixing and matching (like DraftKings, combining What and Why, or Microsoft, combining What and How)
The system, part two: Methods
The second component is the method of your name. Many treat this as a simple aesthetic preference, or try to go down the route that seems easiest based on their capabilities, but it can also be treated as a strategic decision. When determining style, I ask: how many ideas do we need to convey, and how much do we value clarity over flexibility? Can we be intuitive, should we be evocative, or do we have to be cryptic?
Method 1: Intuitive
Intuitive names, ones that audiences can immediately understand with no explanation or thought, are good at conveying a few ideas very clearly and very inflexibly.
These are names like McKinsey, Seven-Eleven, Bank of America, Salesforce, or Yum Brands that hit you over the head with the main message. This is a consideration per market: Hyundai doesn’t mean anything in English, but in Korean it means Modernity. Intuitive names are almost always real words, or light modifications of them.
This is a good approach when you have a clear, consistent, simple idea to build around, which is already captured by existing language.
Method 2: Evocative
Evocative names, ones that audiences can puzzle out with a little thought or background knowledge, are good at conveying a broader range of ideas, somewhat less clearly but very flexibly.
These are names like Maple Leaf, Tesla, Amazon, or Venus that use metaphor and association to convey ideas like Canadian, Electric, Comprehensive, or Women/Beauty. These are often metaphors like Maple Leaf or characters like Tesla or Venus, rare words with little-known meanings like Peloton, words from languages not spoken in your main market like Uber, or coined words cobbled together from familiar roots, like Microsoft or Panasonic. There are a wide range of tactics here, and these tactics can bleed into the other categories— the litmus test here, again, is: could my audience figure out what I mean?
This is a good approach when you have a related cluster of ideas you want to convey, which no existing word says directly.
Method 3: Cryptic
Cryptic names, ones that have no clear meaning until the brand explains it, are potentially able to convey a limitless range of ideas, maximizing flexibility at the cost of clarity. These names are obscure metaphors like Apple or Nike, non-sequiturs like Blizzard, or empty vessels like Pepsi, Lego, Ikea, or L3, to be filled with meaning.
These can include real words, obscure inventions, and initialisms. This is a good approach when you have an unrelated or seemingly contradictory set of ideas you want to convey (in which case the answer might actually be to revisit your brand strategy), and have the marketing budget to shape the meaning you want.
The other important consideration across all three of these name styles is the type of names commonly in use in the category. Fitting in with the same style of name can make your brand feel familiar and easy to understand, while using a radically different style will make you stand out from the crowd, as Apple stands out from stodgy Microsoft.
Recapping the system
Every good naming brief should cover two core ideas: what message do we want to say, and what method do we want to use to say it? Clarity here gets
The Industry’s Weakness: Approach and Construct
I'll level with you, dear reader: everything in that last section was made up. That's how I understand names, and how I work with my clients, and it works for me. But it's not the industry standard. I've used the industry standard, and here's why I don't anymore.
The industry standard system, used by giants like Interbrand and Lexicon, is based on the USPTO's system of categorizing name trademarks. It attempts to blend legal relevance and creative relevance, an admirable goal, but ultimately fails to do either well.
The USPTO system is not concerned with names, but trademarks, based solely on their relation to the "what" message. They categorize potential marks as Generic (Boat, Shoe, Watch), Merely Descriptive (Screenwipe, Breadspread, Mountain Camper), Suggestive (Dri-Foot), Arbitrary (Apple, Old Crow), or Fanciful (Pepsi, Kodak, Exxon). This allows them to determine whether a mark could represent a distinctive offering, or if it’s too literal (either of the first two categories).
This is a great system for identifying whether a term should be trademarked, but it's a terrible basis for your brand strategy, because distinctiveness is only one part of the equation: if all we wanted was to be different, we’d name our companies Asetlkjeant and Lirwejtgkwem. The reason we don’t do that is that we also want to convey meaningful information about the product.
This difference in philosophy is best shown by the USPTO’s use of the word Arbitrary. These can be arbitrarily applied to any product, yes, but they weren't selected arbitrarily, or for arbitrary reasons. More importantly, they only hit one of a brand name's two criteria: whether it's distinctive, but not whether it's informative in the ways a brand might want it to be.
Using legal guidance as the basis for branding
Nonetheless, the naming industry’s standard system attempts to dimensionalize the USPTO system into a two-axis framework, whose most common axes are Approaches (Descriptive, Suggestive, or Abstract) and Constructs (Real, Composite, or Coined).
This was the system we used at Interbrand, and it fails for several reasons.
Message clarity: Neither of these axes gives clarity on what message the name is meant to convey, only describing whether or not it conveys the What message. The system can determine that Apple doesn’t say Computer but would lump it in with something like Salesforce because it doesn’t say CRM. It doesn’t capture what either name does say, or that they’re different. This makes it
Method clarity: Equally importantly, the system also reduces creative considerations to a binary. Real word is a fine category, though the edge cases of obscure terms, technical definitions, and foreign words often muddy the waters. Composite, a full third of this spectrum, account for only about 5%* of modern names, and in my view are functionally identical to real words, not adding any depth to the system. Meanwhile, Coined is such a vast territory that it contains Flickr, Microsoft, and GlaxoSmithKline, far too broad a swath to guide creatives.
Despite this breadth, the two possible categories (dictionary words and invented words) don’t cover the full gamut of naming tactics. Acronyms and human names, which account for about 35%* of company names, are not accounted for whatsoever, at best considered abstract real words (again, placing 3M and McKinsey alongside Apple).
Consistent interpretation: Because these categories are so broad and vague, different namers interpret differently: no doubt any namers reading this article disagree with at least one of the categorizations in the table above. When I worked at Interbrand, I asked our 10 namers to categorize a number of large company names, and almost none of them got consistent results. Coca-Cola stood out as an uninterpretable name, living all over the chart, depending on whether you know what coca and cola are. The descriptive names were the only ones categorized consistently — except by our trademark accuracy.
Legal accuracy: This system repurposes legal terms, creating confusion when the creative process transitions into the legal process. Trademark pros I have worked alongside take particular issue with Descriptive, treated in the USPTO framework as Merely Descriptive: namers tend to lump almost any intuitive term in here, where the USPTO takes an extremely narrow view, and will deny a trademark if they determine it to be merely descriptive. This can really hamstring clients, as their creative agency uses similar terms to their trademark team, but with completely different meanings and intentions, making it hard to tell where the differences lie.
Dethroning the standard
Despite this system’s flaws, it is everywhere. Every naming agency I’ve worked with has used a permutation of it, as have most naming guidelines I’ve seen.
The solution I have tried to use, so far with success, is to approach the beginning of the naming process as a matter of creative strategy: what do you want to say, and how can we best say it? This flows easily from the strategic step that has come before, it helps clients understand what they will get, and feel confident that they understand why it’s right. It makes it easier to sell work up the chain, as the strategy can be easily explained, and the names can be shown to meet those criteria.
I keep this explicitly separate from the legal framework, and when trademark comes up or I work with a client’s trademark pros, I make explicitly clear that the systems are not just different, but for different purposes. Where they aim for legal protection, I aim to start an effective brand.
Whether you use the system I’ve described, or keep using the industry standard for the sake of convenience or familiarity, that’s the one thing I want my peers to take away from this: when doing branding, do branding.
*How do I know how common these names were, and when? I spent the first 2 months of the pandemic researching and categorizing 1200 company names according to my system, developing a database to provide analytical backing to conversations with clients.
Got something to name? Or a whole bunch of somethings to name, and a need for a consistent strategy to name each one? Want a detailed analysis of naming strategies in a given category in a given time period? Want to argue that the approach-construct framework is good, actually?