analysis Startup Names Through A Y-Combinator Filter

MadAboutDomains · Dec 14, 2023

After a thorough read I can confirm that that was a great article, Bob! Thanks again

postscripter · Dec 14, 2023

Bob Hawkes said:
While that seems like an efficient way to get a handle on a topic, is it really? As a student I learned best when I struggled on my own path through a topic, not when someone told me about it.

Sometimes, to truly understand things, it is important to do more than look at a summary compiled through another person’s eyes, and their blind spots and biases, but rather go to the primary data and make your own hypotheses, look for patterns yourself. I ask you in the next section to do exactly that.

couldn't be said better...impressive from the start. Will continue reading now but couldn't resist, sorry.

DrJacoby · Dec 14, 2023

There were several names I liked. For example, TipTap, Upstream and Sweetspot.

freety85 · Dec 14, 2023

Thanks Bob, very informative. I was not using the Y Combinator site previously but I will add it to my bookmarks.
Other sites which I use frequently to get ideas and discover startup name patterns :

startuplister.com
startupranking.com

godeny · Dec 15, 2023

Very inspiring and a great article! Thanks

Naming · Dec 15, 2023

No List?

BrandTello · Dec 15, 2023

Great piece of work Bob

Bob Hawkes · Dec 15, 2023

Naming said:
No List?

Do you mean company list? The link in article takes you to that.

Bob Hawkes said:
I strongly suggest that you browse through recent Y Combinator startups yourself. To do that, simply go to this link which should automatically set you up with the W24 and S23 classes.

Naming · Dec 15, 2023

Bob Hawkes said:
Do you mean company list? The link in article takes you to that.

domain list not companies

Anyway i don't like their names
Magichour.ai
.app
.dev
..
Those aren't serious startups just temporary apps on $1 cheap domains.

I wouldn't invest in startup on notcom domain.

Inframan · Dec 17, 2023

Unless it's an English dictionary word I get the very strong impression most of these start-ups refuse to spend money on a domain.

Pretentious sounding names that infer absolutely nothing.

Rove
camelQA
Shiboleth
Toma
Carma (Couldn't afford the K)
Bilanc
Salvy
Collate (double pretentious, similar to Grok)
Zaymo (rhymes with Venmo)
Celest
Momentic
Duckie (IMO this also wins the childish I'm so Gucci trendy award, look out Duck Duck Go)
Leya
Open Pipe (How is it not possible they don't recognize the XXX connotation?)
Coba
Lingtual
Dili
Trayd (Tray deuce is a slang term, Tray Dee is a rapper, theyd tryd twod hard twod be kewld)
Strada
Smobi

In the world of COD and MMO's these names are META for the most try-hardest git guds.

In all smobiness these names are the James Harden's of tech start-ups. Fakes that file bankruptcy come ROI time.

Inframan · Dec 17, 2023

There's a multi-million dollar AI start-up company. I registered the (.COM). They settled for the (.AI).

At the time I registered the name I had no idea this company existed. I'm also thinking why didn't they register the (.COM) first then register the (.AI).

I've personally registered the singular and plural of several domains. I've also registered the (.COM) and (.NET).

The bottom line is these companies are going to build their brand with any TLD just to avoid buying the exact match (.COM).

Bob Hawkes · Dec 18, 2023

Inframan said:
Pretentious sounding names that infer absolutely nothing.

Thanks for taking the time to look carefully through the names, and share your comments.

I agree with your comment, but I view it almost the opposite way, I think. Yes, I agree, most of the names do not have meaning in the sense that I can look at the name alone, and correctly tell you want the business is. But I see that as a positive, potentially, as it can allow them to change focus somewhat without the need to change name.

Now about pretentious. I agree, but when fighting to secure startup funding and early customers, the business probably has to act a little pretentious, maybe? Someone, I forget who, posted that a great domain name allow you to seem bigger than you are. In a sense, many domain names can help by being a tiny bit pretentious.

As several have commented, a number of these adopted something other than the .com, even when the .com seemed, to us at least, not unreasonably priced. I think that is perhaps one of the most important things to consider. I would suggest the following....

Keep in mind most of these are very early stage. In some cases just a couple of founders, and they were at the idea only stage less than two years ago. True, now they have at least first funding, but when chose name and wrote their white paper they were probably very cash strapped.
This and other looks at startup names suggest that there is a strong preference to single word (dictionary or not) names, not universally and it may contain on location of founder and especially on sector to some degree. To the degree they prefer an alternative extension to a longer word in .com.
I agree with your comment that some seem overly cute. It seems to me some startups like pretentious, some like elegant (not terribly common words) and some like cute and friendly. What most seem not to like is boring service description terms, as were so popular in the 1998-2000 Internet growth.
I wonder if each new wave opens the door a bit wider than the previous wave for alternatives. Like .ai is beyond what we saw with .co or .io or .xyz, it seems. Is that because founders today were seeing founders of a few years ago brand on .io, etc. and that has made something other than .com more accepting? This is just a hypothesis.
Is there any emerging bias against legacy extensions, or is it simply purely a manner of availability and cost. i.e. will some tech startups go in wanting an .io or .ai, or is it just they end up there?

Want a much bigger list of company names? But this time companies that have been tested further in the cycle, and achieved unicorn status? Here is a list from CB Insights of more than 1200 unicorn companies. I had trouble finding even 3 of the 1200+ that had adopted a product or service match name - i.e. a name that when you heard it would unambiguously know what they offered.

Absolutely there is value and a place for product and service match names. Without, or with little, advertising spend you can get traffic to your site.

But as Sten reminded us in this NamePros Blog interview, they don't match the essential purpose of a company name, to be distinctive.

I wish we had a time machine, so we would know how many AI startups will over next 4 years select a name that is a service + AI .com (or AI + service .com) vs service + .ai, vs generic +.ai vs made up / creative + .com, .io. .ai etc.

-Bob

Inframan · Dec 19, 2023

Bob Hawkes said:
I agree with your comment, but I view it almost the opposite way, I think. Yes, I agree, most of the names do not have meaning in the sense that I can look at the name alone, and correctly tell you want the business is. But I see that as a positive, potentially, as it can allow them to change focus somewhat without the need to change name

Now about pretentious. I agree, but when fighting to secure startup funding and early customers, the business probably has to act a little pretentious, maybe? Someone, I forget who, posted that a great domain name allow you to seem bigger than you are.

This and other looks at startup names suggest that there is a strong preference to single word (dictionary or not) names, not universally and it may contain on location of founder and especially on sector to some degree. To the degree they prefer an alternative extension to a longer word in .com.

Is there any emerging bias against legacy extensions, or is it simply purely a manner of availability and cost. i.e. will some tech startups go in wanting an .io or .ai, or is it just they end up there?

Reading your response made me think maybe B2B businesses don't need to be hard wired towards product and service name match. I'm the one that's maybe hard wired into thinking a product or service is your company brand.

Working for tier one automotive suppliers definitely has influenced that perception. Product, Service and Quality are everything.

Because there isn't an obvious indication to what these companies provide and deliberate ambiguous names, they've failed at marketing and branding.

I acknowledge this is the incubation stage, however I need something from the SOC (Standard Occupational Classification) as part of your name. e.g (Liberty Mutual Insurance, Penske Automotive, Del Monte Foods, ect)

I don't dislike all of these names.

Celest and Leya going by the radio test are pleasant sounding.
Momentic is a portmanteau word. I'm not sure what the two root words are. (I did comment on your post when you mentioned portmanteau domain names.) I'm not feeling Lingtual.

An optimistic viewpoint similar to you viewing these names as being versatile. Maybe there's a future opportunity to rebrand towards a more distinctive brandable name. The catch is as other members noted the (.COM) was/is obtainable, but isn't considered necessary.

COM, NET, ORG, GOV represents 99.99% of the websites I visit. I've yet to type (.IO) or (.AI) to find a website.

I'm very new to domaining, but I've already noticed a pattern. Instead of buying
(cloud.com), these start-ups would rather register (kloud.com). Phonetically they both sound the same, that's where the comparison ends. If your company name is an afterthought or insincere my patronage is non-existent.

J.R. · Dec 19, 2023

Viewing the new startups on this list and comparing their domains is always a good exercise.
Often, early startups that see value in the EMD dotcom but can't afford them use (brand+addon & dotcom).

See it a lot with health, telecommunications or cloud companies.

Might be BlueCloud, GoldWireless, etc, but the company and customers refer to the brand as Blue or Gold.

On the list, SimbieHealth .com, Certainly Health or Decoda Health.

Simbie.com is available.

But Certainly.com &Decoda.com are in use by other companies.

In those cases, a future rebrand might be unavoidable unless they want to send traffic to other companies into perpetuity.

Threads.com and clubhouse.com rebrand cases are very recent examples of the consequences of not branding with the EMD dotcom in mind.

Great post Bob.

Ntmt · Dec 20, 2023

I wonder how many of survived Y-startups will rebrand or will secure exact match domain name(s) after some time.
I suspect that looking at early stage startup names may be misleading in this regards

Bob Hawkes · Dec 20, 2023

Ntmt said:
I wonder how many of survived Y-startups will rebrand or will secure exact match domain name(s) after some time.
I suspect that looking at early stage startup names may be misleading in this regards

Good points. Definitely it is a pattern to start with an inexpensive name, and later upgrade when and if the business is successful.

I think looking carefully at the examples are most helpful for those who see selling their names to early-stage startups. I guess it could also be useful to look at the names, but not so much the domains, for those who seek to sell to companies who are upgrading (not these exact names of course, but the type of name that is desired).

In general most startups will fail in not many years. I suspect there are stats re survival rate of Y combinator companies vs startups in general, but did not try to dig that out. I would guess that Y combinator do better, partly because they got the Y combinator incubation experience and learned from that, but also simply because of the funding injections, and the notice they get after being selected. Also, the people at Y Combinator are investing real money, so the fact they researched the ideas and chose to invest in a particular company is some predictor of success. The examples and overall stats I repeated in the article show the spectacular success of some of the past choices.

Thanks again for good points. We should not overly emphasize this relatively small sampling of very early stage startup name choices. But probably best to not ignore it either.

-Bob

Bob Hawkes · Dec 20, 2023

Lots more good points made. Thank you for your contributions @Inframan. A few thoughts below...

Inframan said:
Working for tier one automotive suppliers definitely has influenced that perception.

I think definitely true that different sectors have different name choices, and even needs. Even in this tiny startup set, it stood out to me that medical companies tended to include the specific term health in their name, for example.

Inframan said:
COM, NET, ORG, GOV represents 99.99% of the websites I visit.

I guess depends on location, as well as sectors of interest. Here in Canada I probably visit .ca more than any other extension, although .com a close second, and .org for Wikipedia and a few others. Yes, I know the InternetX annual report suggested .com dominates Canada. Definitely many regions national country codes are strong.

Inframan said:
If your company name is an afterthought or insincere my patronage is non-existent.

I think to be fair, it is not that they are insincere, or think a name is not important. Rather, an early stage startup is essentially proving whether an idea is viable. Yes, a great domain can help them with that. But also, spending say $200,000 to $750,000 on the name they really want, when their total funding is less than that, is simply not an option. True, they often spend more later than they could have acquired the name at the outset, but they need to make sure they have funds to develop their business.

Would you use a search engine that is a misspell of a word that at the time they selected the name only a few math geeks knew? Do they really want us to take them seriously? It seemed to work out for Google.

If you did not know what you know after they became established, what exactly is Uber? I mean it means nothing specific, just a vague superlative. Hey company, choose a name that tells me what you do in the name if you want me to take you seriously! But see, it worked for them, as now we all have come to associate what they do with that word.

We all bring perceptions into what is a good name. Domainers have different perceptions. So do startup funders. So do potential clients. And the people that might fund, or not, those startups. The right name can be a balancing act.

Thanks again for your well-argued and expressed points.

-Bob

Inframan · Dec 23, 2023

Bob Hawkes said:
what exactly is Uber?

When President Obama promoted the GIG economy, he simultaneously endorsed Uber and Lyft. Former White House aids who now work for Uber gave the company access to the Oval Office. That's an unprecedented advertisement boost. Only Super Bowl commercials have a greater impact. And I still don't know what an Uber is.

Bob Hawkes said:
Even in this tiny startup set, it stood out to me that medical companies tended to include the specific term health in their name

Those medical industries that are easily identifiable are useful to the end consumer.

Bob Hawkes said:
True, they often spend more later than they could have acquired the name at the outset

Agreed. As a hedge against inflation maybe some of these start-ups should enter a lease purchase agreement.

pretentious_domain · Dec 24, 2023

Hi Bob,

I'm the Co-founder and CEO of Magic Hour. I found this post because your mention showed up as a backlink to our domain, magichour.ai. Thanks for that.

I didn't know a corner like this existed on the internet. I was pleasantly surprised at how much analysis is being put into naming. I'd imagine there's a large economic incentive involved as well.

I thought your audience might be interested in hearing the naming process of a YC company.

We're just one data point, but I suspect our experience is highly common.

Our company started as an idea in December 2022 to make AI-generated music videos.

I built a prototype, started posting videos on social media, and in January after we started getting tens of thousands of views, we decided to build a waitlist.

To choose a domain, I used ChatGPT to help me come up with ideas.

I created a list in a Google spreadsheet, filtered out any domains above $125 dollars or unavailable, and ranked them 1-5. There were 21 names in the spreadsheet.

Magichour.ai came out on top, and I probably spent a few hours total.

Most of the domains on my list were ".ai" for the simple fact that the ".com" versions were too expensive or weren't available.

I would love a ".com" name, but when you have no revenue, it's not even a thought that crosses your mind to spend thousands or even hundreds for a domain.

At the stage of company we're at, our time is always better spent building your product and talking to users.

Even now, when we have some revenue, users, and funding, naming isn't close to being a priority and likely won't be for many years.

One day, if we reach product-market-fit, that may change and naming may be the highest impact thing we can do, but if that happens, it's a good problem to have.

I suspect a lot of YC companies have a similar thought process. We're all just trying to make our companies succeed, and name isn't going to be what determines that success.

Name is also malleable. I'd estimate up to 25% of companies in this batch will change their name at some point, some more than once. And many will quietly disappear.

When you don't have a lot of customers, changing names just isn't a big deal.

I imagine domain investing is similar to startup investing in many ways.

In the early stage, you have very little signal, so ultimately it's a numbers game.

In the late stage, you will likely have a higher hit rate, but will also have to pay a price accordingly.

Domains sure do seem like an efficient market.

@Naming We aren't a serious startup, just a temporary app on a $1 cheap domain, and you wouldn't invest in a non dot com domain. Thanks for the smile and small dose of inspiration

Let us know if you have any feedback on our product at magichour.ai as well

Bob Hawkes · Dec 24, 2023

Thank you so much for joining NamePros and for so fully sharing your thought process, @pretentious_domain ! Hearing voices from actual startup founders is important if we are to better serve the market for domain names. Very best wishes for every success with your startup.

Here are a few thoughts that your comments triggered for me.

pretentious_domain said:
I thought your audience might be interested in hearing the naming process of a YC company.

We're just one data point, but I suspect our experience is highly common.

I was going to ask whether you feel that the process you outlined was reflective of other early startups, but you already answered that! From your interactions at YC, or elsewhere, do you know if some early startups go the route of using a SquadHelp naming contest, or employing a naming agency, or is that more for later stages in development?

pretentious_domain said:
To choose a domain, I used ChatGPT to help me come up with ideas.

Very interesting that to get the early possibilities you used ChatGPT. If you don't mind sharing, did you directly use ChatGPT or through a product like DomainsGPT?

pretentious_domain said:
At the stage of company we're at, our time is always better spent building your product and talking to users.

I think too many in the domain community are so focused on what a good name can do that they can't see this critical point – that job 1 in the early stage needs to be actually building a product that people want and will use.

pretentious_domain said:
Name is also malleable. I'd estimate up to 25% of companies in this batch will change their name at some point, some more than once.

Interesting. So not simply obtain the exact match name when better funded, but possibly change the actual name itself?

Clearly from your comments you considered .com and .ai, but had a stringent top price level in mind at this stage. Would you be able to share what other TLDs you think early startup founders consider in general? I presume that .io is on the list. Were there others that you considered for your application, like .tv?

Thanks again for sharing your experience, and very best wishes for every success with MagicHour. I really like the connotation of the name.

-Bob

pretentious_domain · Dec 24, 2023

Hi Bob,

Thanks for the reply. You strike me as a Data Scientist/Professor in your approach and I appreciate your balanced perspective.

From your interactions at YC, or elsewhere, do you know if some early startups go the route of using a SquadHelp naming contest, or employing a naming agency, or is that more for later stages in development?

I've personally never heard of any startups talking about their naming process, with the exception being companies that have made it and have created some lore around them, such as Airbnb (Airbnb started off as "Air Bed and Breakfast" because they thought an air bed and breakfast were key parts of their product, and shortened it when that stopped becoming the case).

Because it's not a common topic, I've never heard of SquadHelp or naming agencies. I think the prevailing view is that spending money on choosing a name is not a great use of money or time.

I think if anything, startups will ask a friend, peer, or family member for their thoughts. Note that there's a built-in filtering mechanism in having a co-founder as well, which ensures that at least two people like the name.

Very interesting that to get the early possibilities you used ChatGPT. If you don't mind sharing, did you directly use ChatGPT or through a product like DomainsGPT?

I used ChatGPT directly. I personally don't like using vertical GPTs since ChatGPT is effective at nearly all use cases if you prompt effectively.

I think too many in the domain community are so focused on what a good name can do that they can't see this critical point – that job 1 in the early stage needs to be actually building a product that people want and will use.

I couldn't agree more, but I also understand. When you spend all of your time and energy in one pursuit, you get really good at optimizing, but it's easy to miss the forest for the trees and miss out on insights you could get from engaging with the people involved as counterparties.

Interesting. So not simply obtain the exact match name when better funded, but possibly change the actual name itself?

Yes, exactly. Startups change names for many reasons. Sometime the name you chose is trademarked and you don't find out until later. Sometimes a different name has a better connotation with your users (e.g. Justin.tv renaming to Twitch when they pivoted to gaming).

In each case, changing names is extremely cheap in the grand scheme of things, and users are very adaptable (it didn't kill any of the companies I mentioned and likely made them stronger).

I tend to believe that every company in the current YC batch would be willing to change their name next week if there was a strong and compelling reason for it.

Clearly from your comments you considered .com and .ai, but had a stringent top price level in mind at this stage. Would you be able to share what other TLDs you think early startup founders consider in general? I presume that .io is on the list. Were there others that you considered for your application, like .tv?

In my view, .com is still Tier 1 and .ai is Tier 2.

Beyond that, there's a long list of TLDs like .io, .tv, and .xyz that are more vertical specific.

For example, .xyz is typically associated with crypto and Web 3 startups, and .tv is typically associated with media and streaming businesses.

We would have easily gone with TLDs beyond Tier 2, but it wasn't necessary because .ai had so many cheap, relevant names.

Thanks again for sharing your experience, and very best wishes for every success with MagicHour. I really like the connotation of the name.

Thanks Bob, I appreciate our conversation and thanks for sharing a slice of this world with me. Hope to catch up again once we've made more progress.

Bob Hawkes · Jan 14, 2024

Article Bonus:

I am outside edit window, but here is some bonus content for the article. For the 217 companies from the Summer 2023 group at Y Combinator I did some statistics.

First I looked at the extensions in use, data shown below.

Almost half (47.0%) used .com, and another 24.9% .ai. After that .io was most popular (7.8%) followed by .dev (5.1%) and then .co.

Only 1 used an .org, and none used a .net. .App and .health both had multiple users.

New gTLDs with a single user in the group were .aero (not truly a new extension), .bio, .build, .cloud, .earth, .link, .money, .run, .tech, .travel, .video and .xyz. In total about 12.9% used a new extension.

There were not many country codes in use beyond .ai, .io, .co and .so, but .do, .gg, .la, .re and .uk. each had a single user. None of the startups in this group used well known country codes like .me, .tv and .cc, and national codes such as .de, .us, .ca and many others were also not represented.

I also looked at the lengths of both the company names and the SLD (second level domain) term. Company names averaged 8.2 characters while the SLD average was 8.4 characters.

The number of startups with names longer than 9 characters really dropped off, as the graph below shows.

While many of the startups did use an exact match between company name and domain name in some extension, by no means all. A number used a prefix like try or use, and a few a suffix like hq.

Four of the names used a hyphen, and one was alphanumeric (a school-related service had k12 in their name).

I would think the majority are hand registrations, although did not try to verify this. Clearly a few are on names they would have had to purchase on aftermarket though.

Keep in mind that this is a group of early stage startups, but does provide a glimpse of their thinking with respect to names and company names.

Hope you enjoy this bonus content!

-Bob

GoKaizen · Jan 14, 2024

SH and those Naming agencies do robust business.

pretentious_domain said:
Because it's not a common topic, I've never heard of SquadHelp or naming agencies. I think the prevailing view is that spending money on choosing a name is not a great use of money or time.

So you will switch to .com domain name when you can?

pretentious_domain said:
In my view, .com is still Tier 1 and .ai is Tier 2.

branding · Jan 14, 2024

Bob Hawkes said:
There were not many country codes in use beyond .ai, .io, .co and .so, but .do, .gg, .la, .re and .uk. each had a single user. None of the startups in this group used well known country codes like .me, .tv and .cc, and national codes such as .de, .us, .ca and many others were also not represented.

Which should be a good indication the data used for research is highly geared toward US businesses. They dont represent the majority of the businesses where the money is at.

Cctlds are on the rise. The cctlds mentioned are mostly repurposed and irrelevant long term.

analysis Startup Names Through A Y-Combinator Filter

Top Member

Top Member

Top Member

Established Member

Established Member

Account Closed (Disallowed)

Upgraded Member

Top Member

Account Closed (Disallowed)

Established Member

Established Member

Top Member

Established Member

Established Member

Top Member

Top Member

Top Member

Established Member

Established Member

Top Member

Established Member

Top Member

GorgeousAI.comTop Member

Private InvestorTop Member

Similar threads

We're social

Pinned

Appreciation

Agreement