discuss Pronounceability (Does it pass the radio test?)

[Corey]

The “Radio Test” really is the best way to get a good idea of whether your LLLL.com is pronounceable or not. If your LLLL.com was mentioned on the radio, would listeners be able to spell it without difficulty? It’s better to ask other people than to try and determine this yourself — most of us will be inherently biased towards our own LLLL.coms.

Is passing the radio test still valid today? your thoughts

Cheers
Corey

 

[johnn]

Everything for sale here is pronounceable. You just have to use your tongue wisely.
Don't believe me? Post in the wanted section and you will see what you get from PMs.

 

[hookbox]

The radio test is totally BS now. We are bombarded daily by visual marketing and being memorable is far more important. I can't even remember the last time I listened to the radio. Radio? What the hell is that?

 

[Eric Lyon]

We posted a video for The "Memorable" Domain Test:

 

[NameOmnia]

Actually neither Yahoo or Google pass the radio test. We say they do because we know them as brands; if we did not we would not guess the spelling at the 1st/2nd attempt imo.

Memorable and catchy is what we should me after over radio test etc. We are in a visual era and we are visual beings that's why places like BB are able to sell below average brandables for x,xxx or xx,xxx just by adding a cheapo logo to them.

Think about this for a moment

You see a name posted here in B&W Arial and you go..meh..not good..You see the same name on BB with nice colours and a logo and you go...eff!..why didn't I register that?

 

[gilescoley]

The radio test is totally BS now. We are bombarded daily by visual marketing and being memorable is far more important. I can't even remember the last time I listened to the radio. Radio? What the hell is that?

Its mainly domainers that use the "radio test" phrase, but you only have to ask end-users about this concept and you will see its very relevant and important to companies when picking a domain. Just beause its called the "radio" test, it doesnt mean its on the radio, its more verbal face to face stuff or talking on the phone. Its like saying to someone "Go and have a look at my website wwww.shonky.com and whether you would need to spell it to them or would they go away and type it in correctly first or second time..or would they struggle to get to the site at all..or would you feel the need to spell it out to them after mentioning the site. Thats all the "radio" test means, its not about the radio.

Actually neither Yahoo or Google pass the radio test. We say they do because we know them as brands; if we did not we would not guess the spelling at the 1st/2nd attempt imo.

Memorable and catchy is what we should me after over radio test etc. We are in a visual era and we are visual beings that's why places like BB are able to sell below average brandables for x,xxx or xx,xxx just by adding a cheapo logo to them.

Think about this for a moment

You see a name posted here in B&W Arial and you go..meh..not good..You see the same name on BB with nice colours and a logo and you go...eff!..why didn't I register that?

Not sure how you figure they dont pas the radio test? I know I would probably spell them both correctly first time if I had never heard of them before.

I know most domainers dont bother with logos, good names sell themselves. I dod know a lot of new domainers who struggle to sell their names, will add a logo to try and make it more appealing, and most time it wont work. Im not saying it never works, it certainly will work on occasion, but it will be primarily the domain that will sell and not because of the logo.

If a company doesnt like the look or sound of a name, a colorful logo wont convince them to buy it. Personally, I love a good logo, but I have tried this in the past and had no luck at all, and I know a lot of other domaienrs who have tried as well...Just my 2 cents worth

 

[carob]

If anyone has ideas on what makes a word pass or fail the radio test, I'll be happy to try to automate.

It's essentially whether the user can successfully transcribe what they hear and the transcription exactly matches the written domain name.

In the best case there should be no alternate spellings the user could try. Think bong.com or dog.com.

A user with mastery of phonetic symbols can transcribe anything accurately - but the problem arises because English allows you to represent the same sound in writing in different way.

You could rank things by the number of phonetic variants they risk: so splunk.com loses a point because k is replaceable by c. It might lose fractions of points because within that variation there are sub-choices: c or ck.

Variants for available deliberate mis-spellings could exist - using 2 for to/two or omitting e before r as in Flickr. IMO those are - as people have said here - very dependent on being seen written so you get a graphic memory. People's memories work differently - we know some are more responsive to graphic or audio inputs.

A weaker version of the radio test is memorability - can the user recall the sounds? At least then they can use search to find the item and can mention it to others. So a radio test score could return a basket of ratings - audibility, memorability, number of syllables, and also the feel of the word. Is splunk attractive?

All the more reason to use emojiz for domains.

 

[aramyus]

@carob

If I understand well, you suggest 2 criteria for passing 'easily' the radio test:

- no alternative spellings for each syllabs/subsets of letters
- memorable

For now, I can't think of a way to measure memorability (afterall: ImproveYourCreditRatingNow.com is memorable)
It also reinforces me into the idea that a pronounceability index may be a good lead: if a word is composed on frequently used syllabs/groups of letters, it will more easily pass the radio test

Anyhow, thanks for your suggestions

 

[Asfas1000]

Context (and proper English) also helps with the radio test.
Anyone would spell "cell phone" vs "sell phone" although the sound is the same.

 

[carob]

I don't see a formal radio test returning a yes/no result, more of a ranking. And it could be part of a family of tests.

The radio test itself could have different factors and the user could decide which is most important - correct spelling, existing dictionary word, pronounceablity, uniqueness, number of syllables, pattern, price, availability, similar usage by others... the list could get a bit long so yes you need to narrow it down to something practical, and be clear what you are excluding.

 

[aramyus]

Radio test is definitely a better indicator to measure the attractiveness of a domain than pronounceability

There is no clearcut border between a pronounceable and a non pronounceable word/domainname (examples include tumblr, trst for trieste and more generally many words with foreign origin or abbreviations)

I've done some research on pronounceability. I thought that if a word was composed of known syllabs/diphtongs/groups of letters, it would be automatically pronounceable and vice versa. But I found many domains that do not obbey these rules and are reasonably pronounceable. For example: afpot is pronounceable (at least I can pronounce it), yet, there are no english word that contains the group of letters afp. Similarly, tlaro or axtix can be pronounced eventhough there are no subsets of english words that can be easily related. Incidentally, my examples would have a hard time to pass the radio test.


Probably a pronounceability index (from 0 to 100% rather than a flag pronounceable/non pronounceable) taking in account how frequently it appears within english words could be a better alternative. I am not too sure if it would relate simply to a radio test.

If anyone has ideas on what makes a word pass or fail the radio test, I'll be happy to try to automate.

 

[gilescoley]

@Silentptnr

Or Googil, Googel, Googill etc...I prefer Yahoo but there are short names out there even better than both of them when it comes to the radio test

 

[Corey]

I'm really enjoying hearing everyone's valued opinion, much appreciated folks.

Cheers
Corey

 

[Eckhart]

@carob

if a word is composed on frequently used syllabs/groups of letters, it will more easily pass the radio test

That seems like a good starting point for building a script that measures pronounceability, although you should check some of the previous algorithms on this matter. For example using it on 5-letter strings would be great for us domainers and we could adapt this code. Although in practice I got decent results just using pronounceability patterns filtering random 5L's. But if you come up with something let me know, maybe I could add something to it.

From an automation perspective, the focus should be on pronounceability instead of passing the radio test (this would also eliminate some unnecessary rules) because, for example, funy.com doesn't pass radio test (you'll have to also say it's just a single N) but it's pronounced with ease and it makes a great brandable nonetheless, so a script needs to rate it high regardless of failing the radio test.

 

[Asfas1000]

Being pronounceable and passing the radio test is not exactly the same thing.

 

[gilescoley]

The radio test is very important when it comes to short domains, especially when selling to end-users, it makes a domain for brandable and desirable when its hard to spell incorrectly. Yahoo is a great example of passing the radio test with flying colours

 

[NameOmnia]

Exactly, but 3 ways is not a lot,

3 or more attempts in the brandable world are a lot and those are typically names that require a lot of marketing and advertising to be recognized and stick in people's head and that's where marketing agencies play their role with the strategic use of colours, shapes ( and slogans etc )

Not even BB according to their rules (rules which they break all the time but that's another story ) would list a name that takes more than 1-2 attempts.

From BB site

"What kind of names do you accept for listing?

To accept or not accept a domain is often a tough decision. Many very good and "brandable" names are submitted, but because we try to keep the marketplace small enough for a single buyer to browse in an evening, we've created some basic guidelines:

We are looking primarily for non-keyword, short and catchy names. For these types of names, we mainly focus on:

• If the name is based on a dictionary word -- and sounds the same as that word -- it should have no more than one spelling variation or "error" away from the original word. This makes it easy to explain to customers. For example: Digg is "dig with two g's".
• Pronunciation should be straightforward.
• Spelling should be as expected.
  • Ask your friends to spell your name -- can they get it in one or two tries? "

At BB 40,000 domains must mean a small marketplace that can be browsed in one evening.. >>> funny side note <<<

 

[NameOmnia]

and the ones that do sell are primarily to fellow domainers

Come on Giles..now I am not a fan of BB either but to affirm that BB is not a good example when it's the number 1 site for that type of names ( whether we like it or not ) and to affirm that the majority of names are sold to domainers ( also considering the price tags ) it is to say the least incorrect.
You know it, I know it, we all know it.

 

[aramyus]

adapt this code

I adapted this code, but I was not happy with the results.

I took a different approach for pronounceability: I first built a table of decomposed english word into groups of letters. My separators are diphtongs/voyels and groups of consonants. I do the same for a given domain, and I check if each subset of letters can be found in my table of decomposed english words. If yes, I consider the word as pronounceable, and vice versa.

If I take your example funy, I decompose it into f/ fu/ fun/ uny/ ny/ y.
I notice that each group is used in an other english word and hence it's pronounceable.

However, some subsets are not that common. For example uny is used in only 6 english words that include puny, unyielding, bunyan. None of these words are very common.

On the contrary, unny (for funny) is used more often and in more common english words such as sunnyvale, runny, gunny

This thread's discussions gave me the idea to build a pronounceability index that would take in account the frequency a group of letters is used in english and the frequency at which these english words are used in the english books.

I assume 'funy' will have a much smaller pronounceability index than 'funny', and this index could be used to have an idea of how easily it would pass the radio test.

I know it's not as straightforward as it can seem. One big problem I face are brands that have become part of our common vocabulary (think IBM, realtor, tumblr, flickr...) Clearly, IBM is not pronounceable, yet it would pass the radio test. IBMer would probably pass it too, and it's not really prounounceable either.

 

[Eckhart]

I'm not sure it does fail the radio test

If you tell someone to go to funy.com the obvious assumption will be funny.com, so I say it doesn't pass the radio test, but 'cough' doesn't have a similar sounding word and will pass radio test because people will assume it's indeed cough... the problem would be with kough.com. Anyway, my example may not be the strongest and probably we could find a better one, but ultimately the point is there are plenty of names that fail radio test and still sound great.

However, some subsets are not that common

I see your point. The frequency should be just one of the factors of the index and you could add others that will influence the score, such as.. if it also matches a pattern such as CVCV or VCVC it gets higher score... if the 2 vowels are the same it gets better score (more melodic).. if the vowel is A it gets better score than if the vowel is U.. and so on, we can refine it with more rules.

 

[aramyus]

Radiotest, pronounceability, memorability... are not a guarantee of sellability.

For me, they are just indicators, among others, to detect if a domain is worth or not more investigation.

For now, if I had to choose between Radiotest and pronounceable, I'd choose radiotest.

But I am sure we can find many flaws and many exceptions in each indicator.

 

[jamesau]

The 'radio test' is really just a cleaner way of asking 'how easy is it to understand on the phone'. A lot of business is still done on the phone, so it's a relevant an important test.

Audio advertising does still have a role but less so radio. More likely podcasts.

If you're looking for something that truly passes the radio test you shouldn't be going for made up terms. It should be solid dictionary words. eg. 'Good Website Design' not 'Gooweb'.

 

[Silentptnr]

The radio test is very important when it comes to short domains, especially when selling to end-users, it makes a domain for brandable and desirable when its hard to spell incorrectly. Yahoo is a great example of passing the radio test with flying colours

Actually, it doesn't really pass the radio test. Google is a better example of passing with flying colors.

Yahoo could be interpreted as YA WHO. Google is GOOGLE no matter what.

 

[Silentptnr]

Maybe its why yahoo got beaten by google. Google passed the radio test better.

 

[NameOmnia]

Not sure how you figure they dont pas the radio test? I know I would probably spell them both correctly first time if I had never heard of them before.

Interestingly enough 2 comments ago you said this

Or Googil, Googel, Googill etc..

Given that Google is an alternative spelling of “googol” I guess it is a bit of a stretch to think that you would guess the spelling at the first attempt without having heard it before; same applies to Yahoo like someone else has pointed out..but hey..if you are a genius good for you.

I know most domainers dont bother with logos, good names sell themselves. I dod know a lot of new domainers who struggle to sell their names, will add a logo to try and make it more appealing, and most time it wont work. Im not saying it never works, it certainly will work on occasion, but it will be primarily the domain that will sell and not because of the logo.

If a company doesnt like the look or sound of a name, a colorful logo wont convince them to buy it. Personally, I love a good logo, but I have tried this in the past and had no luck at all, and I know a lot of other domaienrs who have tried as well...Just my 2 cents worth

"good domains sell themselves" .... except that the definition of good domain names becomes much more grey than what we all would like it to be once we remove the obvious great and category defining domain names. Daily sales prove it.

If adding logos hasn't worked for you it does not mean it does not work on a general level and I would not dismiss it as the thing that new domainers with crappy domains do; million dollar marketplaces that specialize in brandable / company names wouldn't have made "the look" of the names such a major aspect of their selling otherwise.