One man's trashy is another man's treasure

The long long tail of creators and platforms


Cameo is a company that has taken advantage of the long long tail, realising that a small gesture from a minor celeb can make a big difference to a fan.

1. Cameo and the long long tail

Kevin Kelly, cofounder of Wired magazine, coined the phrase 1,000 true fans many years ago. Li Jin recently built on that idea with her 100 true fans article. If you haven't read either, the gist is that an achievable amount of loyal fans (100 or 1,000) are enough for creators to make a living. For example, someone who had 1,000 fans paying $100 per year would have enough to get by in most cities; likewise for a creator who had 100 fans paying $1,000 per year. And getting 1,000 fans doesn't sound too far-fetched.

Both of those articles attest that the internet has cut costs for creators, enabling a long tail of talent to be discovered and supported. We're all weird, and it's now quicker, cheaper, and easier to find like-minded people producing stuff we want to see. No longer will your virtuosity in acting like a dog remain undiscovered; it might even make you a six figure income [1].

Whether most creators in this new economy can actually hit that 1,000 true fan goal remains to be seen. We will see many success stories of independent artists, but I feel that we'll fail to see all the people who dropped out along the way. If you've ever tried to grow your follower count on social media, you know that it's more difficult than it looks. In this day, anybody can be a star, but not everybody will be a star.

I'll hop on the bandwagon of writers naming frameworks, and call this the long long tail effect, seeing as the only precedence for the term is a children's book. What the long long tail effect says is that long tail is longer and weirder than people expect, even if they already expect the strange. What it implies is that if you want to pick a winner in this new economy, you're better off picking a platform, rather than choosing a creator. Although the chance of success for any single creator is low, the chance of success for the platform consolidating creators is high.

Hence, rather than discuss depressing dog acting markets, today I want to look at the platforms that power them. In particular, I want to take a closer look at Cameo, a marketplace that provides personalised celebrity video messages. For the cheap price of $30, you can get a custom video message from Kirpa Sudick, Kevin Fortenberry, or three (!!!) videos from Paris the mini pig [2]. People normally get them to wish family and friends happy birthday, graduation, or some other celebration, to the delight of the recipients.

I first mentioned Cameo about a year ago while trying to guilt people into buying me a Jenna Coleman cameo [3], and they're a good example of a beneficiary of the long long tail. The marketplace works because the individual return for these celebrities on their own is low, but the combined return of all of them is high. Paris the mini pig on her own is not likely to make enough for her owner, but combine her with Fiona the Hippo or Lola the Sloth and you might actually have a viable business for the platform providing the service.

Marketplaces succeed by aggregating demand and supply, so let's start by analysing Cameo's.

Cameo's demand and supply show product market fit

Cameo doesn't seem to have a company motto, and what it should be is "One man's trashy is another man's treasure." Yes, I know they'll never officially use it, but it comes as a direct consequence of the long long tail. You may not care about contestant #23 on the Asian Bachelor, [4] but someone somewhere out there does. A lot. It would make their day, and they're willing to pay.

This asymmetry in user experience is why Cameo is in such a great spot. Customer delight? Creating minimum viable happiness?" If you buy Aunt May a cameo from her favourite movie star, that's all she'll talk about for years, and maybe she'll finally stop bugging you about getting married.

I'd assume that these cameos are simple for celebs to film as well, since all they have to do is read off a script and stare into their phone for a few minutes. Given how easy it is, no wonder so many have joined. Since the talent also sets the price, that implies that the celebs are charging whatever they feel gives them a balance of bookings and billings. They could raise their rates if they were overbooked, and lower them otherwise. Efficient market in action here.

A quick glance at the site shows prices mostly vary from $30 to $100, which isn't surprising to me. If you've ever been to a comic con before, photo ops or autographs usually go for somewhere in that range, implying there's proven demand at that price point. What Cameo has done is enable the long long tail. Now you can get a comic con substitute experience anytime, anywhere. By Grabthar's hammer, what a savings.

An additional incentive for celebs to be on Cameo is the brand building opportunity for themselves. They not only make money from the video, but remain in the consciousness of their true fans. The biggest fear with fame is to be forgotten, and Cameo curtails that concern. Celeb fans are incentivised to share the video too, making them more likely to resurge in conversations. As Cameo puts it, the talent is getting paid to promote their own content.

Supply also brings more supply. When you see your fellow Bachelorette star (that got eliminated earlier than you) getting a viral video reaction from her fans [5], you're probably slightly jealous. When you learn that she also got paid for that 1 minute video, you'll likely sign up for the platform yourself. It doesn't remove the need for Cameo's sales team to sign up celebs, but does make their job easier [6].

Is that enough though? Cameo got off the ground by working the long long tail, and will continue needing the support of that group for a long time to come. However, it makes sense that they'd want to start targeting bigger stars as well, due to the higher prices that can be charged and the larger publicity. Rather than harming the long long tail of celebs, this actually helps them. They can now say that they're on the same platform as So-and-So, a point of prestige. The fans of the long long tail are also searching specifically for them, so more celebs shouldn't cannibalise the earnings of others already on the platform.

Can they reach a tipping point where more A-listers feel comfortable joining? I'd guess this is like the discomfort of AirBnB or Uber when they were getting started, and that the idea will normalise over time. They've already gotten people like Elijah Wood (Famous for Lord of the Rings), Sarah Jessica Parker (Famous for Sex and the City), or Snoop Dogg (Famous for smoking weed) on the platform, so it doesn't feel that much of a stretch to continue expanding there.

We've established the case for demand, and the case for supply, seeing that there's clear product market fit. What about the economics of the platform itself?

Cameo has a high take rate, semi-scalable cost structure, and large addressable market

To better understand the platform economics, I'd want to look at three things [7]:

  • The take rate, or fee that Cameo charges

  • The likely cost structure of Cameo

  • Total addressable market size (TAM)

Cameo has a 75/25 revenue split; Cameo takes 25% of every $1 booked. That's a relatively high take rate, and from cursory research higher than what celeb agents get (10%) or what celeb managers get (20%), though of course the services provided are different. I'd prefer for them to lower their take rate, but that's always been a personal bias of mine when thinking about marketplaces.

Could an agent or talent set this up themselves, and disintermediate Cameo? Potentially. Celebs could have made offers on their Instagram or Snapchat accounts all this time, but most have not done so. It's likely a combination of privacy, social norms, and hassle. The celeb doesn't want to give up their personal contact info, implying the shoutout has to be done on the platform. However, that means that the celeb's feed will be cluttered with cameo videos, which is probably not desirable. Having to set up all the payments and track orders is also annoying.

Does Cameo offer an easier way of earning money without having to advertise yourself? Definitely. Once they're signed up, celebs don't need to do anything else other than get the order and shoot the video on their cell phones. Promotion, privacy, payments - all of that pain is handled by Cameo.

Cameo's cost structure does concern me. Since the product is delivered digitally, they likely have low marginal costs, a couple of percentage points in payment processing, and some unknown but low amount in data infrastructure [8]. That's great, and what I'm more curious about is the semi-fixed cost base.

About a year ago [9], Cameo claimed to have 120 employees, of which 80 were dealing with talent. That's a lot of people doing either sales or support roles. It also shows the problem with serving that long long tail - there's just too many of them. You'd have to assume that this cost function grows slower than the growth rate of talent, in order for the business to be viable. You might be able to outsource customer support, but talent support is probably still going to be done in house.

Yelp and Angi homeservices are other companies that also cater to the long long tail - Yelp selling ads to restaurants, and Angi being a marketplace for home service contractors. If Cameo ends up having a similar cost structure, that implies ~50% of their revenue is going to end up in sales and marketing [10]. That's large, and could be the key line item deciding whether Cameo is profitable or not.

Lastly, how large is the market? Market sizing is usually more art than science, especially when you start doing top down modelling. If you took the entire $700bn US entertainment market and said Cameo could take a few basis points of that, that's already a billion in revenue, though on highly aggressive assumptions. Given Cameo's raised venture funding at a $300mm valuation, I'd assume some VC has done the math to justify them eventually hitting a unicorn valuation.

I do feel that the above math assumes too large a base, so let's look at what comic cons do instead. Supposedly NY Comic Con does $100mm in revenue, so it seems reasonable to assume Cameo could one day get the same amount of reach as a single city convention. That's roughly 1% of the US population paying $30 for a cameo, which doesn't seem that crazy to me. You'd probably get a smaller percentage of the population, but a higher average amount paid.

Whether such business is recurring or more one-time in nature is also something to think about. I think that there's enough interest for cameos to be ordered

Overall, the take rate seems lucrative, the cost structure potentially concerning, and the US TAM reasonable to support a sizeable valuation for Cameo.

Cameo for the future

If I were Cameo, here's some things that would concern me for the future:

  • Copycat competition, from new or existing social media

  • Cameo quality

  • Expanding beyond the US

As mentioned earlier, it does seem Insta or Snap missed this market trend, due to wanting to preserve the existing user feed experience. People will pay for even more interaction with their celebs though, as seen by virtual gifting in China, Twitch video game streaming, or OnlyFans in the US.

Given that Insta and Snap both have ecommerce options, it seems possible for them to also launch a cameo feature. You'd go to a celeb's page to request a cameo, get that cameo back as either a public or private video, and be able to share that with friends, all on the same platform. Since dedicated fans probably already follow their favourite celebs on social, that's less friction for them compared to going to Cameo's site. What Cameo can do here is to make it the easiest experience for celebs to do cameos, and continue growing such that they give the most bookings.

Cameo doesn't suffer from the demand/supply matching problem that most marketplaces face. Since users know what they're looking for, you don't need sophisticated user rating or recommendation systems to match talent to the demand. That's a potential cost savings that can be spent acquiring more talent instead.

The lack of user ratings could be an issue, depending on the quality of cameos as the business scales. Some celebs do better than others in sticking to the script. I'm sure Aunt May would be happy regardless, but probably slightly less so when she hears the wrong name mentioned her happy birthday video. However, I'm unsure if the talent likes to see ratings or comments on their Cameo profile, given the ego hit.

LL note - I’ve since realised Cameo does do user ratings, and edited the paragraph above.

Cameo doesn't seem to have effective sorting options at the moment either, so that's a potential new feature. Sorting by most expensive or most requested could probably increase total bookings. The concern again is probably how the talent is going to feel when they can see their relative popularity. There's a reason why Masterclass doesn't release information on how popular their classes are.

I’m unsure if making a user profile to book Cameos is worth the trouble. Sure, you can link it with Facebook to make creating an account easy, but it seems like doing guest accounts would work as well. I suppose there’s some benefit in suggesting more cameos that user might like, based on the previous ones they did.

International expansion seems like an obvious next step, with Europe likely the most natural next geography. The difficulty is probably setting up the sales team in that location and getting the first few local celebs on board. I don't see major localisation issues here, but could be mistaken. Unlike some marketplaces which only have local network effects, Cameo seems to have broader network effects, since getting a UK celeb will generate demand from a US customer that watches british shows. To some extent, that will also benefit expanding in non-English speaking countries, as there will likely still be demand for international celeb cameos from the local population. Lotta avengers fans in China.

Cameo concluding thoughts

If you've seen one of those celebrity quarantine house memes, Cameo is the product that brings that to life. The regular fan now has the ability to buy some time from their favourite celeb.

Cameo found initial success through the long long tail, realising that a small gesture from a minor celebrity could have outsized impact on an individual. There's clearly demand for the product, and supply is probably constraining the marketplace growth for now. Their TAM seems large, take rate could be lowered, and cost structure is potentially an issue. Looking to the future, international expansion seems like an obvious next step, as long as they can retain their product quality and not see copycats.

2. & 3. The long long tail in finance and personal life

I normally also do a finance piece and a personal growth piece. Given the length of the tech section above and the GPT-3 article on AI from last week, I'm going to keep this month's section on both brief.

The long long tail is also a phenomenon in finance, and the easiest example of this is Robinhood, the startup offering commission-free trading. Opinion is split on whether this is a good thing or whether Robinhood is stealing from the poor to feed the rich through payment for order flow [11]. Regardless, the rise of Robinhood and competitors has enabled a newer, larger, weirder group of traders to grow. And just like before, I'd pick the platform to be the winner rather than any single individual.

And what the long long tail implies for your personal life is that regardless of your interest, you're likely to be able to find a group that welcomes you. Especially in this more isolated time, finding a small community for yourself will make you feel less alone, whether that's some niche subreddit or discord chat group. You're a unique snowflake, just like everyone else.


  1. I spoke to the cofounder of Baru, which is a marketplace for custom furniture. They're raising a round if you're interested.

  2. Avenues is partnering with the Long Now Foundation to offer an online gap year program designed to accelerate the development of future inventors and entrepreneurs.


  1. The Tesseract: Economics of an Information Space and how networks define the digital world

  2. "This is a few short stories about things that never change in a world that never stops changing."

  3. Varieties of Motte and Bailey

  4. Using algorithms to improve your touch typing

  5. I played around with the chat bot Replika a few years ago, seems like they've made some improvements and did a relaunch since then.


  1. I did warn that it would be weird. Also, it seems like she makes six figures a month. 1,000 true onlyfans indeed. Sigh.

  2. Nope, I have no idea who any of these people or pets are, but I assume some people absolutely adore them. Paris drives a mean bargain.

  3. To no avail, sadly. One can hope. I’d settle for David Tennant.

  4. This is not a real thing, in case people were wondering

  5. Or is it him? I never remember which one has all the guys and which one has all the gals.

  6. Cameo also has a talent to talent referral system, where talent gets a 5% share (out of Cameo's take) of the bookings for the new referred talent.

  7. Technically TAM isn't part of economics, but I've grouped it here for convenience

  8. I know some people dislike the net revenue treatment and would say the 75% cut is COGS, but that doesn't really matter to the cost structure below that line, which is more important imo.

  9. Can I just say how annoyed I am whenever articles don't have an obvious "published on" date?

  10. Based on Yelp's filings here or ANGI's filings here. I can't remember if they account for support costs within the S&M line item or outside of it, so that 50% number might be even higher.

  11. I personally think the concerns are unwarranted, see Matt Levine or Patrick McKenzie for more detail