I’ve seen a lot more sweetshops around recently. Specifically, these Victorian-themed tuckshops and sweet connoisseurs, plucked from two hundred years prior and plopped in the middle of a high street that doesn’t quite know how to welcome its new tenant. One of the sweet shops near me plasters ‘Toys, Trains, Treats” on dated storefront panelling; it’s the shop you thought your grandparents went to when they were young, except the steam trains on sale are battery powered.
This wasn’t just a Baader–Meinhof phenomenon, where upon seeing something once you perceive it everywhere. During the first year of the pandemic, an extra 73 sweetshops popped up in the UK, net, according to the Local Data Company. That’s not a huge number, but it’s pretty significant given the broader lockdown context; the total number of clothing shops fell by 1,236 in the same period. The shop near me was set up in September ‘21. It begs the question: what’s so special about sweet shops?
I had spent my Saturday morning at Waterstones buying some books, at the sweet shop questioning the shop assistant, and now I’m writing about it on Substack. In days gone by, I would have bought my physical items at Sainsbury’s and, in comparison to today, where I have my Substack and my main website, all my digital publishing would be through the same technology stack, probably Wordpress.
Sweets are nice, but the real difference is the existence of a shop that will focus all of its resources on their specific niche product. I bought some weird caramel chocolate flakes that defy description, but they were excellent. I certainly paid a premium, but next time I want sweets, I’ll go there again, because they know their product; I just bought what the shop assistant recommended. At Waterstone’s, I picked up Train Dreams because I told the manager I was after contemporary fiction; I use Substack because it’s built for writers. What makes these businesses special is that they serve niche markets in which they can give people much deeper value.
Bundling and unbundling are not new concepts. It’s a pretty simple convection model: you start with a business, over time you add more products, services, and features, and then you get bloated, do everything ‘kinda well’, until a niche competitor comes along and does one of your services really really well, and bit by bit you get pieced apart. The complexity convection is a pattern that cannot be unseen.
People have observed this happening to our contemporary digital landscape for about a decade, beginning in practice much earlier. Craigslist was unbundled into many things, including Reddit, which is now being unbundled into many more things, like Nextdoor. In the language of a16z, the famed tech investors: platforms get unbundled into verticals, which turn into platforms, at which point the cycle resumes, and all of a sudden I’ve turned into a strategy writer.
I’m not the first to herald this unbundling in the physical world either. Peter Day wrote an excellent essay predicting “the heartbeat economy, almost a concierge approach to customers,” in 2013:
“This would be a worryingly intimate relationship for many businesses, but one which might provide a new kind of rewarding marketplace if they dared to be far closer to their customers' individuality.”
He wrote about it in the BBC, but today he would write on his own blog.
My real contribution is to note that this challenges why unbundling and bundling happens at all. The main idea behind why we sway between bundled and unbundled products is the underlying technology. Marc Andreesen, co-founder of a16z and of Mosaic, the first web browser, talks through the concept of the newspaper:
“The newspaper bundle, the idea of this slug of news and sports scores and classifieds and stock quotes that arrives once a day was a consequence of the printing plant, of the distribution network for newspapers using trucks and newsstands, and the famous newspaper delivery boy. That newspaper bundle was based on the distribution technology of a time and place.
When the distribution technology changed with the internet, there was going to be the great unwind, and then the great rebundle, in the form of Google and Facebook and Twitter.”
This is certainly true: we could only put live events on Twitch after network infrastructure made it cheap enough to broadcast live video around the world.
But it’s no coincidence that physical businesses are unbundling at the same time as digital ones, despite relying on different technologies, and often very little tech at all. Neither technology nor innovations built upon it come from magic: what consumers actually want plays a key role; today it’s driven by the niche market vibe shift.
Why? Well first, consumer demand fosters technological innovation. I took a cursory glance at solar panels last month, where growing consumer demand and growing production have induced new technologies; if people hadn’t bothered to use streaming services in the early days, there wouldn’t have been the investment that made them possible at scale. Substack was built because its founders believed writers wanted to offer readers deeper value in a specific niche - and that readers would pay for it – not due to any new technologies, just due to the niche market vibe shift.
Second, even when this technological innovation happens, there is no rule commanding consumers to use every new use case that’s suddenly made possible. Google, Facebook, and Twitter all had to create new bundles that people actually wanted to use: do Google+, Google Answers, or Google Notebook ring a bell?
I think the reason I’m seeing more sweetshops around is because of a niche market vibe shift. The technology for serving more niche communities has always existed, but now consumers actually want it. But what does this mean?
I have two quick points before we move on: why this insight is important and why this shift is happening.
First, understanding that bundling and unbundling cycles sway partly due to social preferences, as well as technology, is important to understanding where these sways stop and start, and how to take advantage of them. If it’s true that people want more meaningful, participatory communities, that tells us a lot more about where the future is going than simply observing the entire possibility space opened up by new technologies. When I looked at those failed Google products, I got the impression that Google - founded by Stanford grads - originally sought to be a lot more academic than it is today, with tools for professional research and note-taking. We can understand that these products failed because they weren’t aligned with the niche market vibe shift of the 21st century, even though they were technically feasible.
Second, my quick take is that this shift has its roots in society’s search for reciprocal connection. The 21st century is probably the only century that will ever be worth calling ‘the internet century’ (by the 22nd the internet won’t be exciting anymore). Today’s internet is very uni-directional, with big influencers megaphoning to their followers. But people also want to belong to something, and as I wrote a few months ago, building connections within your community makes it so much more fulfilling. You feel you’re really participating, and this is really why I go to the sweet shop. Society just invented the internet, and is searching for meaning. Also G.F.C.
The rest of this essay is to explore how this shift is influencing culture. We can think pretty big about all the unbundling opportunities new technologies bring: in Life After Google, Gilder points out that even money could be unbundled, though I don’t think this is on the cards. It’s possible, but consumers don’t actually want to unbundle money into a store of value, unit of account, and so forth. Very few communities are built simply around money as a tool; everyone will not become currency traders; a niche can’t be built around something that everyone uses, and most people don’t care.
I want to instead focus on culture, which people care a lot about, and is interesting for many further reasons. It interests me how great cultural contributions both capture some image of society whilst also moving it forward. Culture is also beautifully adaptable and flexible and fluid, changing rapidly and exhibiting immense diversity of contribution and consumption: it’s a collection of niches. I expect this vibe shift to play out in culture very soon. (It goes without saying that this will occur on the internet.)
That idea of culture as a collection of niches deserves interrogating: culture is also something that we all share. And sure, there will continue to be big Instagram accounts and large supermarkets; shows and books that everyone talks about and events that unite the masses. But as Rex Woodbury notes: “Celebrities like (Tom) Cruise are a dying breed”. The 2020s will be defined by the “long tail of creators”, each creating en mass for far smaller communities with much deeper engagement. In the 21st century, we will seek experiences that feel, in some important ways (and not all), totally and emphatically human — that’s why people love eSports; we’re in the moment with the players. For the hobbies and niches that we love, for which we’re far more than mere followers, we’ll find people serving that niche with dedicated communities built around them.
Late Checkout is an agency that doesn’t deserve the negative tag associated with such companies; they capture the situation perfectly:
“The old way of building products was to try and be everything to everyone. Products were chunky, bloated, and didn’t serve the individual or communities.”
“People want to feel a connection to the products they use. They want the internet to feel human and alive again.”
It’s exciting to see people building the future, and culturally we’re going to see exactly this: content will be built around niche and dedicated communities with genuine human connection8.
In 2008, Kevin Kelly wrote one of the most seminal contributions in culture tech. The founding editor-in-chief of Wired wrote that, in the internet age:
“To make a living as a craftsperson, photographer, musician, designer, author, animator, app maker, entrepreneur, or inventor you need only thousands of true fans.”
A thousand dedicated fans who love your niche. Sound familiar?
So what took so long?
It’s certainly the case that the technology of the era made this difficult. Kelly hoped that the internet would be the ultimate matchmaker. Bring the creators and the consumers online, and they will find each other. To an extent, they did, especially in the early days of the internet: in 2008, writing about 1000 True Fans looked sensible. But it was still pretty hard: you could publish your own content, but without mass-market services like Facebook, Instagram, or YouTube, that helped you discover, it was tough to find or be found.
The introduction of these services, monoliths to be, helped in one sense: it was a lot easier for people to find you on the internet. But the price for this matchmaking was almost everything: Facebook and Instagram take almost 100% of all the revenue creators see on their platforms; YouTube takes 45%; Spotify paid an average of merely £700 to each artist last year. These monoliths do the matching, but make it virtually unaffordable to create in the first place.
It was a vibe shift, not a technological one, that struck the first blow to this impoverished paradigm. Twitch, Kickstarter, Patreon, Substack, amongst many others, are all services built for creators to cater for the dedicated fans of their niche; all built amidst the community-centric vibe shift born in the 2010s. Well before the 2020 crypto bull run, a16z heralded a 100 True Fans model in 2020, arguing that this was made possible with even more niche services from Podia, Teachable, and Kajabi. (These companies all help people sell educational courses, which undermines the argument to an extent, but you only need 250 fans paying £20/$25 per month to earn ~3x the average UK salary or $75,000, so you get the picture.) With the alignment of technology and the vibe, the impoverished paradigm of community-stricken creatorship is collapsing.
You’re reading on Mirror, so you probably know all this; you could easily skip to the next phrase in bold.
In the early 2020s, new technologies like blockchain, distributed ledgers, and cryptocurrencies really came in vogue. A distributed ledger is triple-entry bookkeeping. Double-entry bookkeeping was invented in medieval Italy: two parties to a transaction would both record it, giving you more data about your business and more verifiable figures, making it easier to trade and pursue enterprise.
Eight centuries later, distributed ledgers are triple-entry bookkeeping: as well as the seller and buyer, every participant in the economy makes a record of the transaction, and every participant is using the same, enormous, book. Blockchain is the technology of how you add to this book and run software with it, and cryptocurrency is one use case: the ledger essentially records how much currency you own, which everyone verifies. Obviously I’m simplifying, but I’ll refer to these technologies as ‘web3 techs’.
The significance of blockchain becomes clear upon interrogating why Facebook, YouTube, et al. can take such high take rates on creators. One side is discovery: you rarely find creators on Patreon. This happens, still, on big social networks. Everyone is on these platforms (well, not Facebook) and when you consume on these platforms frequently, they’re experts in figuring out what content you’ll click on next; big data, Markov chains, etc. There are some problems with this, mainly that you get served addictive content rather than niches that you want to engage deeply in, but generally they do this job pretty well, and web3 techs don’t hugely change this.
But there are a lot of platforms for discovery — this isn’t the competitive moat it once was. People use Instagram, Snapchat, TikTok, Pinterest, and much more, in addition to the discovery platforms we’ve called out above; as mentioned, most people have moved beyond Facebook. Creators could just move to whichever discovery network takes the smallest cut of their revenue. Apart from one thing: they can’t. Everything you’ve ever uploaded to Spotify and all the followers you’ve earnt on Insta: the discovery platform manages this, and you can’t take your community with you. So it’s impossible to leave without starting all over again. And hence discovery networks squeeze creators for revenue.
And hence, the game-changer. Because just like with your cryptocurrency, you can store your writing, your music, your art, your videos, your followers, on a distributed ledger, or “on chain”. As LBRY puts it: “LBRY does to publishing what Bitcoin did to money.” Just like with your currency, everyone participating in the economy agrees that “yes, Leo, you wrote this blog”. Sites like Spotify and Instagram don’t manage your songs and your followers because these songs and follower relationships are stored on chain, and Spotify is reduced to one of many organisations competing to merely display this content in the best way, and help you discover new content as well. Creators are not beholden to any of them – and neither are consumers.
So now the community economy is fully on the way because finally, dedicated participants in a niche community can meaningfully support their creators. When people are deeply engaged in something they care about, even the original consumers become contributors, and we’ll start to see vibrant ecosystems formed around niches and creators. Greg Isenberg explores the Loot ecosystem, an NFT project upon which creators have realised myriad different ideas; Jenkins the Valet is a great example of community-led storytelling, which will become a category-defining cultural genre in the 2020s, on par with ‘art’ and ‘literature’. Jadyn Violet, a musician, told me that “I’ve never got this much interaction on my art before.” He's been spending his time building new ideas to engage his community, spending time speaking with his fans, and making more music — this is what his fans love him for.
Business people often talk about the Minimum Viable Product your business needs before you start selling. The MVP isn’t dead, but all businesses (not just creators) will soon require a Minimum Viable Community to make a mark in their market.
What will underlie all of these communities are tokens of membership, probably tiered, which tie a community together. They’ll be non-fungible, which means they’ll be unique and personalisable, thus allowing for deeper engagement. Consumers get dedicated community and deeper value in a niche they care about; creators are finally rewarded fairly for their work.
In 1999, Excite was a $6.7bn search engine, but collapsed soon after the 2001 dotcom crash. Google was only just getting started, but after aligning with the niche market vibe shift of the 21st century, became the favoured search engine of billions.
Joe Kraus, the founder of Excite, explains why: Excite had been a 20th century company, targeting revenue from the top ten American companies, as media businesses had been doing for decades. Meanwhile, Google focused on attracting advertisers — and millions of them. Why did this work? “The 20th century was about dozens of markets of millions of consumers. The 21st century is about millions of markets of dozens of consumers.”
Hopefully that didn’t actually feel like 3,000 words; I’d love it if you were to subscribe for free or mint one of 5 editions.
Image credits: HelloDorking; Kevin Huynh; Peter Yang
Mirror doesn’t do footnote links well yet, but I couldn’t bear to publish without them.
1 I don’t have the percentage figures for the change in sweetshops and clothing stores, I’m sorry.
3 Stratechery anticipated this for (cable) TV in 2013 and watched it happen in 2017: TV used to keep us informed, educate us, show live events, tell stories, and provide escapism. But now we get our information from Google; our education from YouTube, Coursera, or Masterworks; our live events from Twitch; our story-telling from Netflix; and our escapism from social media. It’s not as neat as this of course: in reality, everything just exploded into a plethora of mainly mobile options; but unbundling is happening.
4 I number crunched some of the Instagram follower stats here and would estimate that of all the follower-followee relationships, the largest 1% of accounts receive about 50% of all following relationships.
5 I’m undecided and unclear on what this means for ‘great’ or ‘classic’ cultural contributions. If a great cultural contribution captures an image of society, but also moves it on, how can it be classic, if being classic is continuing to capture that image of society: hasn’t society moved on and been influenced by your contribution?
6 Woodbury is mainly making the argument that it is specifically “a 20th-century culture centered around white leading men” that is dying, but whilst it’s true that celebrities are becoming more diverse, what’s more important is that they’re becoming less relevant. The social media era is dying; the social community era is coming.
7 It’s ironic that in principle ‘agency’ is the most important thing in the world, whilst ‘agencies’ that help brands and creators reach their fans are typically seen as vultures.
8 I also expect niche and tailored content that is emphatically artificial: a further implication of this, from the unrelated and sufficient cause of AI too, is that individuals will asymptote towards consuming cultural content that is tailored just for them: an audience of one. As artificial intelligence (which was also covered [last month](http://The asymptote we reach is that for most people, most of the time, their reading and viewing material is created just for them)) gets more able to accurately predict our interests and preferences at a subtle scale, we get to a world in which content is designed – no, generated – to appeal specifically to us. GPT-3 is an OpenAI algorithm that can conduct incredibly lifelike conversation (at times). The Next Web covers the story of one developer who used the algorithm to sell bots that spoke with the individual personalities of an individual chosen by a customer, including dead relatives. As always, there are challenges that accompany change, as well as opportunities; AI will continue to develop, but the risk and threat of AI trained to stimulate and manipulate us need to be managed. Further reading: The Diff explores this idea from a more financial perspective; the NPR illustrates a perfect example, with fake LinkedIn profiles generated explicitly to pique your interest; Tyler Cowen touches on the issue from the other perspective, that of the person using the bot.
10 Take the price of Bitcoin as a proxy for ‘web3 in vogue’ for a moment. When a16z published the 100 True Fans articles in February 2020, Bitcoin was worth under <$10k, and wouldn’t get above that price until October. The niche market vibe shift predates the cultural significance of web3 and crypto.
11 I’m simplifying a lot; this isn’t a technical introduction to web3. Most obviously, it’s not the case that each person has to manually agree everything in a distributed ledger. Rather, it’s every computer server; and it’s not every computer server, it’s those who proactively commit to contributing to manage the ledger and communicating with it. A good introductory podcast here; a more detailed introduction here; a more technical and key whitepaper here; and for more depth I suggest the further reading listed under the whitepaper.
12 This illustrates Vitalik Buterin’s (founder of Ethereum) infamous Uber quote: taxi profiles will be stored on chain, and Uber, Lyft, and more will compete to show you the same drivers, giving the drivers much more power and compensation — which they need. (There we go, I actually got a politics link in.)
13 This idea comes from Alexis Ohanian, co-founder of Reddit. I don’t think MVC will replace MVP - you still need a good product - but your community has to be part of the same conversation.