Loading summary
Transcript1 lines
- [00:12]
Rob Meyerson
Welcome to How Brands Are Built, where branding professionals get into the details of what they do and how they do it. I'm your host, Rob Meyerson. Thanks for listening. Today's episode of How Brands Are Built is a special mini episode about the Brand Names Report. The Brand Names Report is something I put out on February 29, 2024, and you can download it for free at howbrands are built.comBrandnamesReport on today's episode, I'd like to walk you through some of the findings from that report. Just a quick summary of what's in there. And this is a video episode, so if you're listening to this, you might want to tune into YouTube so that you can see the pages of the report as I walk through them. But if you're just listening, I will do my best to talk through what's on the screen as I walk through the report. Let's dive in. Here we are in the Brand names report, dated February 2024, a little overview of what's in here. So there's an intro, there's some information about the data, a lot of caveats that I'll get into, then an executive summary followed by an overview of the more detailed findings. And that's about as far as I'll go in this summary walkthrough. So introduction. I won't read through what's on the screen right now, but just hit a couple of high notes. Just what is this report? So I was curious about some polls that are on a site called Brand New. Brand New is the leading brand identity review site. It's edited and written by Armin Vitamin, and it's mostly about visual identity design. But of course, a lot of times when there's a rebrand from a logo or visual identity standpoint, there's also a new name or if it's a completely new brand, then it needs a new name. And so occasionally, not consistently, unfortunately, but occasionally, the blog will have a poll at the bottom of one of those posts asking users whether they think the name is fine or great or bad. Those are the only three options. Is it, is it great, fine or bad? And so I saw that these polls were happening and I just thought there must be a lot of data there about these names. Let's analyze it and see if it tells us anything. And so From July of 2020 to June of 2023, visitors to the site cast a total of 22,000, over 22,000 votes across 131 brand names. And so that's what is in this report. And then I wanted to see if I could find any kind of interesting trends or insights into why some names might perform better than others. Now let's get into those caveats about the data first. It all comes from that one blog, almost all of it. I'll explain where some of the other data comes from. Like I said, not every new name post on the site has, has a poll, unfortunately, right now. And since I think August of 2020, brand new, has been a subscribers only site. And so the only people voting are subscribers, which would lead to the question of who are these people? A lot of them, I'm sure, are designers. But I don't think we can reasonably argue that this is a representative sample of the target audience for these brands or really anything else that you would want. If you are really going to analyze how quote unquote a name was. It's just, it's just whoever the subscribers are to the site, some of the names didn't get many votes at all. And so I'm a little worried that it, it's hard, you know, it's not the best data. If you only have 20, 20 votes on a name, how seriously can you really take that? It's almost so few that the agency that came up with the name or did the design work could probably sway that pretty heavily in one direction or another if they just asked all their employees to vote. So not, not accusing anyone of anything. But you know, when there aren't a whole lot of votes, I get a little less comfortable analyzing that data. And so I have tried to mitigate that in a few cases and only look at names with say 100 or more votes or something like that. The other data that I used in here, there's some larger data sets. So I scraped together some really long lists of unique names. Brand names from all over the world got up to about 10,000 once I had deduplicated those lists. And so I don't have any votes or any kind of data on, you know, how people have perceived those names. But in places in this report where I just wanted to see how long is the average brand name and things like that, I use that larger data set. I also pulled together just some data about the English language that comes from an analysis by Wolfram Research. And there's a link in the report if you want to learn more about it, as well as some research conducted by Peter Norvig, a researcher who was at Google or maybe still is at Google, and He analyzed almost 100,000 distinct words from books Scanned by Google. So again, just gives us some sense of the English language that we can then compare against these names and you'll see what I mean when we get there. There's a footnote here. Brand new has over 30,000 subscribers. So if you wanted that number, total votes on the names range from over 2000 for one of the names to just 10 for one of the names. And I will get into sort of who these people are. One of the fun things I did in this report is I sent it to some great namers and asked for their opinion on it. And the opinions, as you might guess, were kind of mixed. And so Anthony Shore, chief operative at Operative Words, a fantastic namer. Google him and you'll find some really great articles about him in Forbes and the New York Times and some videos of talks he's given. But he says at the end of the day, the brand new naming polls are really beauty contests judged by people largely outside the brand's target audiences. Who cares what they think? So fair enough, Anthony. But he did like the report overall, but was very blunt about that criticism which I already mentioned. You know, these, these are not the people buying from these brands. This is not even just representative, I don't think we can say, of the general public in, you know, any given country or the world. And so he's really concerned about, you know, biases and, and sort of lack of relevance for this, for this group. And I've, I've peppered quotes from Neymars throughout just to add a little more color commentary to what we're seeing in the report. Okay, so who, who are those 30,000 subscribers? So I don't have data on who voted necessarily, but I have data from Brand New on who those subscribers are. It's about a 50, 50 split. Male, female, about 60% of them are 18 to 34 years old. Over 70% are English speakers. Chinese and Spanish are come next, but it's really overwhelmingly English speakers. And you know, it looks like the visitors to this site skew a little bit younger than those on the Internet overall, but otherwise it's a pretty good, pretty healthy, I think, mix of people. All right, so the executive summary, which I think is just this page. I mentioned already 131 names over a 36 month span running from 2020 to 2023. And that's 22,769 individual votes. So what did we find? First off, the most preferred names, four out of the five are two syllables long. They're either real or coined, meaning invented, made up words and they are suggestive or abstract meaning. These are not descriptive names. They are. Some people call them evocative or metaphorical, or if they're coined, they might even be empty vessel names. If you don't know these terms, you can go to howbrands are built.com and look at the glossary to learn how at least I define them. But they're not pure descriptive names. And also four out of the five are six characters or less, which is really short. So that is impressive. And there's a nice consistency there to four out of the five top names. But I keep saying four out of five because the outlier that fifth is a long descriptive name. It's the Avocado Collective. And we'll talk about that name a little bit more as we dive into the data. Shorter names seem to be preferred in general, especially when it's a coin name. Real word names and maybe compound names, which are two words shoved together like Zipcar are slightly preferred over coined. Abbreviated names are the least preferred versus those other name types I mentioned. Real, compound or coined found some things about the length of the names. So looking at that larger data set of over 10,000 names, most of them are between 5 and 10 characters long. Fewer than 10% are over 14 characters really long names. In the larger data set, 52% of the names are a single word, 37% of them are two words, meaning there's a space somewhere in the brand name. And 11% of them are three or more words, which is actually a little higher than I might have thought. The median coined name is slightly shorter than the median real word or compound name. I think that makes sense. We'll talk about it more as we as we dive in. I looked at the initial letter of brand names, frankly. Partly it's a really easy analysis to do once you've got all this data in a spreadsheet. So over 40% of names begin with the letters C as in Cat, A P as in Peter, or S as in Samuel. Frequency of that first letter loosely tracks with what you find in the English language. So yes, it's C, A, P and S. But that's not that surprising given how frequently just English words at least start with those same letters. And I'll show you that data in a little bit here. And then last thing is new name versus rename. It seems that entirely new names are preferred over renamings, and that names that result from mergers and acquisitions when two brands are coming together and they either need to sort of fuse those names or create an entirely new name. Those are the least preferred on average. So overview of findings. Here are all the names. If you're listening, I'm not going to read them all out, but we have a really interesting mix here. They're ordered here in terms of how many votes they got, so that that one that got over 2,000 votes was actually the Seattle Kraken, a hockey team. But then we have some things that may be familiar to you. Meta Stellantis More sports teams Cleveland Guardians, Washington Commanders Pearl Milling Company the new name for what was an offensive an offensively named pancake syrup and sort of breakfast brand, all the way down to things that got very, very few names. And so there's just a big mix in here. So looking at the most and least preferred names overall, I already told you a lot about the most preferred names. I mentioned the Avocado Collective. That's actually the number one name. It's a new name, it's a rename for Advanced Packing and Marketing Services. It got a 79% score. And what I mean by score when I say that it's just a little algorithm we came up with. And when I say we, that's me and Alex Foss, a researcher formerly of Adobe. He's now at LinkedIn for how to take the great, fine and bad votes and sort of fuse them into a single number. If you're interested in the math, it is in this report, I think somewhere in a footnote, but 79% is the highest that any name got and that's the Avocado collective going. The second most preferred name got the same score, 79%. That's also a new name, a renaming for Scott's Cheap Flights. So worth mentioning here that both of these shortened the name considerably from what it used to be and went from something really descriptive and dry to still descriptive in the case of Avocado Collective, but sort of less so and certainly less dry. And then going is just a suggestive name from Scott's Jeep flights. So I do think, and this is mentioned at the end of the report, which I'm probably not going to get to in this summary, that when you're renaming something, a lot of how people react to the new name depends on what the old name was. For better or worse, the third name is Izzy I Z Z I. Fourth is Milo and in fifth place is revealing for these great names. I tried to list who created them where I could and so worth mentioning that both Milo and Reveal in fourth and fifth are from Ragged Edge, an agency in the uk. So congratulations, Ragged Edge on creating some great names over the past three years. The mostly bad names, Pearl Milling Co. Is near the bottom. So I'm going sort of in descending order here. So the last one I mentioned will be the one that came in dead last in this report. Pearl Mill & Co. Was fifth to last with a score of negative 71%. I probably should have mentioned that earlier. That score that I talked about goes from positive 100 theoretically to negative 100%. And this is negative 71%. It had 24 great votes and 454 bad votes. So that's how we got to a negative by doing the subtraction there and looking at the percentage of the total votes. It's a real word, descriptive name. It's obscure in terms of meaning. It's very long. It was Created in House 128th out of 131was EUSPA, an abbreviation. 129was Serendipians, a coined name, very long and hard to pronounce. Next one I'm not sure how to pronounce. Tellore's network of engaged universities came in second to last. And then dead last is ITA Airways, a new name for Alitalia. And I mentioned this earlier. I think part of the reason people hate that is because they liked Alitalia. And there's some anecdotal evidence for that in articles that were written around this name at the time. So it's not just that ITA Airways is pretty dull, but also that it's replacing a bit a brand that a lot of people liked. Another way of showing the scores is to look at the great, fine and bad votes on sort of a bar chart so that you can really see just visually how many votes each, how many votes of each type each name got. So this page shows 32 names with 200 total votes or more. There's a pretty wide range. Also maybe worth noting here that you can kind of tell how polarizing names are by how many great and bad votes they got. And so there are some names in here where there aren't that many fine votes. Not, not as many people coming down in the middle, but people either loving or hating it. And so it's interesting to point to look at those and think about why they might have been more polarizing. My friend Helen Gould, a fantastic namer here in the San Francisco Bay Area, mentions that whether the general public likes a name is typically unimportant. What counts is the target audience's response and whether the name does what it needs to do. Many names now deemed highly successful. Looking at you, Google were initially seen as very weird. Novelty makes people uncomfortable, but it is that very novelty that can set a new standard. So very well said from Helen, and similar to Anthony, maybe doubting the importance of getting reactions from a sort of general audience on Brand New, but that is the data that we have and so that is what we've analyzed. All right, I mentioned C, A, P and S were the most common first letters, followed by E as in Eric and M as in Mary. Over 40% of the names begin with C, A, P or S. You can see if you're watching the video or if you tune in later, how that tracks against the English language. And you will see that in most cases where we see spikes with letters in brand names, it's the same in the English language, but there are some notable departures. So X is 25 times more likely in brand names to be the first letter than it is to be the first letter of an English word. Y is over five times more likely. H and D are the reverse. They're more likely in English than in brand names. I did the same analysis with that larger data set and really just confirmed what we saw with the 131 names from brand new. B, as in Bravo, shows up a little bit more here. It's 8% of brand names start with that. Biggest departure still are X is one of them, but also Z and K, which are four times more likely in brand names. And then with the exception of A, which a lot of brand names start with, vowels, are always more frequent in the English language as initial letters than they are as initial letters in brand names. So for some reason, E, I, O, U may be not as popular at the beginning of brand names as they are in the English language and certainly not as popular as A or some of the consonants that that I've mentioned. So just a thought from me here. I think it's not that surprising to see brand names, especially when they're coined over indexing on initial letters like X and Z and maybe K as well. Brand owners can choose any word they want, right? They can choose Zenith that starts with Z because it sounds cool. And they can invent entirely new words like Xerox or pull from Greek, as is in the case of Xerox. And so these letters, they may be seen as more distinctive and memorable because they are rare in the English language. So you would expect to see this effect of the more rare letters in the English language, the more likely they are in the beginnings of brand names. That rarity is evidenced by the fact that they are worth so many points in Scrabble. Right? X and Q and Z. And to that point, I'm a little surprised we don't see a lot more Q Q brand names, which in this data set at least, there wasn't a huge difference between brand names starting with Q and English words starting with Q. All right, looking at length, over half the names are one word and six characters or fewer. Did the same analysis with the larger data set. Length skews a little bit longer here when we look at more brand names. Also just worth noting that in general, the brand names do seem to skew a little bit longer. I think that's because on average, a single English word is going to be larger than a data set that includes brand names, which, as mentioned, some of them are two or even three words or longer. So that may be part of the reason we see this idea that they're skewing longer. Shorter names are preferred just based on a correlation analysis. It's a weak effect, but we do see the effect. Some really, really long and really unpopular names like Telluray's network of engaged universities. I mentioned another one, Victorian Gambling of Casino Control Commission, super long and really disliked by the voters on brand new. And so that is part of what's causing the effect. But then we also have some of those great names that were only six characters long or five characters, like Going and Reveal. So if you look at the report or if you're looking at it on screen right now, you can see some of those names called out and get a sense of why that correlation is the way it is. So the effect isn't too strong. But a preference for shorter names matches the conventional wisdom of namers. This has always been conventional wisdom. It's been measured by academics. A 2012 study found that companies with short, easy to pronounce names have higher valuations. And then simplicity, which is related to length, also seems to make a difference. So research from 2021 suggests that when consumers want a product they can control, and in this case, they did a study on a golf ball, which obviously you want to be able to control where that goes. Brands should avoid using difficult to pronounce product names. They found an effect for, you know, a relationship between difficulty in pronunciation and just preference for that brand. So from here, the the report goes into findings broken down by construct. So looking at real words versus coined words and compounds and abbreviations, I'm not going to get into the weeds here on this podcast episode, but I will just say really quickly, kind of an overview. Most of the names in the report are either real words or coins. So that's where we have by far the most data. A quote here from Scott Milano, founder of a naming firm called Tange, based in New York. He says we see clients react more positively to real or coin names over compounds, both during the exploration phase and with the final names they launch. I won't say compound names are totally dead, but they're definitely not popular right now. So that is borne out by this data. Over the past three years. Real word names and maybe compounds, although we have less data on them, seem slightly preferred based on the mean of the voting scores on those different names. But there's a pretty wide range. People loved and hated some of each type of name across the board. Real word names have the widest range of length. Coined words all skewed shorter, which again, because you're able to make up the name, why would you make it long? Part of the reason, maybe for coining a name, is to shave a few letters off of it, like taking the E out of Flickr or something like that. And so not that surprising that they're skewing shorter. From here the report just goes into real word names. Sort of the same analysis, but just specific to real word names. We still see that effect for shorter getting better votes. Same thing for coined names. It's an even stronger effect there. Here I'd say it feels like a real correlation. Shorter names perform better. Compound names, we only have seven in the report, so it's a little harder to conduct much of an analysis. But I've just put the same pages in here so you can see them. And then also looking at 16 abbreviated names, things like acronyms or alphanumeric names in the report, mostly these were not liked, but still wanted to share the data. Hard to tell whether shorter is better here because they are really short. And then just one other thing I'll share, I mentioned at the top, renaming seems to be less preferred than new names. New names seem to be given more leeway. And the mean of new names in the report is 28%, which is not that bad. Whereas for renames it's minus 5% and for mergers it's minus 28%. So pretty strong looking effect there. And then when we are replacing names, again, it's not a very strong correlation. But if you're going to be replacing a name, it does seem that shortening it if you can, seems to be the right way to go in terms of at least what we see in this data. I'll read a couple more quotes to you. Ken Pasternak, a great namer and a friend of mine, chief strategy officer at an agency called 2x4. He says naming factors such as length, memorability, meaning and relatability matter tremendously. But the number one success factor in a name is the commitment of the organization to its name and how it uses it. The greater the commitment, the greater the acceptance among the audiences it is meant to engage over time. That said, longer, more obscure names and those devoid of emotion may take longer to take hold, which means more unwavering commitment from the brands or organizations supporting them. And then Armin I wanted to quote him in here because he runs Brand New. He says naming is one of the most difficult aspects of creating or reshaping a brand. The process is as much about creativity as it is about compromise, whether it be for legal matters with the trademark officer or emotional matters with the clients. And what this analysis demonstrates, referring to this the Brand Names Report, perhaps infuriatingly, is that there is no right or wrong approach or solution, nor a hint of what makes a name good or bad. And therein lies the thrill of one day getting a name just right. So I think he's right. Arm is right. There's no right or wrong approach necessarily, but there are hopefully some hints of what makes some names, maybe not good or bad, but more or less popular when you subject it to a popularity contest like the polls on Brand New. And hopefully some of that has been clarified for you in this Brand Names Report. So that wraps up this little summary of the Brand Names Report. Again, if you want to download a copy for yourself as a PDF, it's absolutely free. You can just go to howbrands are built.com brand names report and let me know if you have any trouble finding it or getting it, or if you have questions about it that I haven't answered on this podcast episode. Hope you've enjoyed it. Thanks again for listening to How Brands Are Built. If you like this episode, please leave a rating and a review wherever you get your podcasts. How Brands Are Built is a production of Heirloom Agency, Inc. Our theme music is by Esha Erskine Project. I'm Rob Meyerson and I'll talk to you next time.