Where did the long tail go? (2022)
honest-broker.com
honest-broker.com
The first major example is about the availability of TV shows and movies on Netflix, but this isn't a reflection of consumption at all, or even of Netflix. Movies went down because other studios pulled their movie content. But it also shows that TV shows quadrupled from 530 to 2,108. That's big-time long tail increase.
The next example is about more sequels in theaters, but theaters have never exhibited long-tail characteristics to begin with, so that's irrelevant.
Then it goes on to Spotify and admits that it hasn't reduced its long tail, but says he "wouldn't be surprised". That's not evidence -- that's nothing. Then he claims that long-tail artists can't make a living on Spotify -- but the long tail has never been about content creators making money, that would be nonsensical in the long tail. It's always been about the platform making money from it.
Then he finally moves to a fast-food restaurant and discovers that loyal customers make up most of revenue. Which, once again, has nothing to do with "long tail" because the long tail is about products, not customers.
He finally concludes:
> We live in a Short Tail society. And it’s getting shorter all the time.
Yet provides zero evidence for this, and it's easily refutable just by looking at television. We used to all watch the same shows, now we all watch different shows. It's common to mention your favorite show and not only has your friend/coworker not scene it, they've never even heard of it. Same with your favorite band, same with your favorite author. Total available content has exploded. We live in a Long Tail society.
Same with mainstream movies + 4 broadcast networks + a few premium cable channels compared to all that plus all the streaming stuff, all the stuff on Youtube, etc. Hardly anything gets the population share of viewers that something like I Love Lucy or Mash or Friends used to get.
Indie video games, blogs, self-published novels, the list goes on and on and on and on.
So not about whether there were 530 TV shows that were increased to 2,108, but whether the distribution of watching is wider or more like a power law.
The author doesn't provide any evidence that the decades-long trend of "more options = broader distributed viewership" has broken, and the person you're replying to has shown that the "more options" trend is largely continuing. The author is cherry-picking also-not-relative-distribution numbers from a relatively small time frame and extrapolating wildly.
Let's look at the longer timeframe.
From Wikipedia for I Love Lucy:
"The episode "Lucy Goes to the Hospital", which first aired on Monday, January 19, 1953, garnered a then-record rating of 71.7, meaning that 71.7% of all households with television sets were tuned to the program, the equivalent of some 44 million viewers.[71] That record is surpassed only by Elvis Presley's first of three appearances on The Ed Sullivan Show, which aired on September 9, 1956 (82.6% share, 60.710 million viewers and a 57.1 rating ).[71] The overall rating of 67.3 for the entire 1952 season of I Love Lucy continues to be the highest average rating for any single season of a TV show.[72]"
MASH: "The series premiered in the US on September 17, 1972, and ended on February 28, 1983, with the finale, showcased as a television film, titled "Goodbye, Farewell and Amen", becoming the most-watched and highest-rated single television episode in US television history at the time, with a record-breaking 125 million viewers (60.2 rating and 77 share)"
By the time we get to something like Friends we're down to 52M live viewers for its finale, and it's only shrunk since then. Netflix doesn't give equivalent counts or US-only ones, as far as I can find, just "at least two minutes within N weeks" type stuff and similar; I believe if they had that level of cultural ubiquity, they'd be shouting it to the rooftops more explicitly.
And if they did, the linked article would be citing it. But instead, they're just speculating that all the numbers and trends we do have better measured are wrong.
If you look at ticket sales counts instead of box office for movies you'll see a similar large representation of older stuff, and that's a quantity comparison even with a smaller population to boot .
Remember that we started with a ~dozen TV networks showing one show at a time. The concentrating effect was so immense that the viewership numbers from the 70s and 80s have never been surpassed.
Take a list of the highest-viewed programs of all-time: https://en.wikipedia.org/wiki/List_of_most_watched_televisio...
The only modern programs that even rank is literally the Super Bowl. The last shows to break 100mm viewers was in 1977 and 1983.
For primetime programming things peaked in the 80s with the Cosby Show, with ~30m households tuning in every single night for a single TV show. Viewership of linear broadcast programs has been on a slow downwards slide since.
And that's not because people are watching less programming overall - but because they're now watching a vast array of other programs, including non-traditional programming like YouTube.
The tail has never been longer. Not only can you find a much wider variety of content than ever before, but the long tail is immensely popular! People are making a living doing shows with a few tens of thousands of fans, and major high-profile, high-budget productions get nowhere close to the viewership from TV's heyday (Succession's finale had ~3m viewers!) because the distribution of viewership is so much wider than before!
Lol, a dozen? There were ABC, NBC and CBS that might (or might not) have all been available in your area. Plus PBS if you counted that. I didn't like Masterpiece Theatre, so just 3 for me.
The largest cities might have had a few independent channels (not networks), but for most of the country, that was it.
not exactly, it is also about narrowing of people's interests based on the wider range of opportunities.
If the wider range of opportunities means that the people can now stop wasting their time on Sexy nurse stories and Voracious Space Aliens by narrowing in on their true interests - Sexy Nurses that marry voracious space aliens to exclusion of other stories - but if instead what happens is a wider range of media being consumed without narrowing in on very specific sub categories then what has happened is not a long tail but simply consumers expanding their range of choice.
Both of these phenomena - long tail and expansion of choice would lead to limits of viewership because there is in either case more media fighting over the viewership but while the result is pretty much the same the causes are slightly different.
For example, in October 2022 Youtube on TV matched Netflix's July 2022 record number of 8% TV viewing hours. That is all long tail.
https://www.nielsen.com/insights/2022/sports-provide-a-lift-...
Yesterday, I discovered a musician on Spotify who has just 3,000 monthly listeners and 300 YouTube subs. The bulk of my news comes from substacks and small YouTube channels. Even my entertainment is spread across half a dozen platforms big and small.
That was the sales pitch of this book, which - so far as I know - made the phrase popular.
https://www.amazon.com/Long-Tail-Future-Business-Selling/dp/...
What actually happened is various platforms moved to monopolise distribution. Only some of those are true open markets with low access and marketing costs.
There are always a few exceptions. Like the TikTokers - always cute/hot - who make millions producing lowest-common-denominator content. And some writers on Amazon who make millions producing... lowest-common-denominator content.
The premise of the book was that creators of all kinds of content - including those making exotic, experimental, and generally not-lowest-common-denominator work would find it easier to make a living.
There's some great original music on Spotify and Bandcamp, but it's almost impossible for anyone to find it.
There are quite a few niche YT channels in the 25k -> 100k range who are getting by, but most of them only make any money from sponsorships and product placements.
The real cost of entry isn't access to a distribution platform, it's access to expensive marketing, advertising, and sometimes sponsorship.
That's what publishers of all kinds used to pay for - and still do, but only for TV/movies and the top music headliners.
Everyone else doesn't get access to that kind of reach and influence.
If you're marketing a cold project you can throw some money at Amazon and Facebook ads and try to get some TikTok virality. But producing that extra content and keeping track of impressions and ROI is practically a full-time job in itself and requires a much broader skill set than just being a good creator and requires spare cash to burn on paid marketing that may be a complete failure.
The 2008 book promised that good content/products would pretty much sell themselves. All you had to do was find your niche with a bit of Googling and start selling into it.
That very much isn't what happened.
Isn’t this a pretty good definition of the long tail though? They are able to sustain making content for a niche audience, driven by access to a platform that enables distribution that will allow them to reach that audience who would not otherwise be able to find the content at a scale that would make sponsors interested.
I am very interested in outsider/experimental/extreme music. It has NEVER been more accessible.
I am very interested in a rather niche subcategory of tea. It has NEVER been more accessible.
Art house cinema, indie videogames, bespoke accoutrements... you get where I'm going with this.
This author is the reason the Long Tail is not flourishing. This author is content with Netflix and Spotify. This author is the problem.
That's orthogonal to a long tail discussion though, which is about the distribution of streams (how many views/streams go to a small part of the catalogue, and how many to the long tail), not their mere availability.
It's especially hard for me to believe when the pre-modern-internet-platform availability of that long-tail content was zero. The percentage of views/listens going to that content literally had nowhere to go but up.
First, TFA author might very well be wrong (or not give numbers to prove himself right), but I wasn't supporting the author.
I was clearing up what someone could imply from the parent comment that just because obscure music "has NEVER been more accessible" this necessarily means we have a long tail effect.
Your point shows the same misunderstanding in a more explicit way: "The percentage of views/listens going to that content literally had nowhere to go but up", still doesn't automatically translate to long tail consumption. If it has gone from being 0% to being 20% of the content consumption for example, the "long still" would still be nowhere competitive with the fat powerlaw "head" (the stuff still getting 80%).
Was it only ever claimed that it would eliminate the head?
It doesn't have to flatten the curve entirely to "expand the scope of the arts" or "nurture alternative voices" or "get indie films to larger audiences" or let "music be more surprising" (paraphrasing from the claimed benefits as restated by the author).
Or, if you move to the author's economic claims, "make a lot of money selling to these microscopically small groups of customers"... I would take more issue with that in general, but without a better defined "lots" we don't need Etsy to make its sellers more money than Walmart makes theirs - we just need them to make a "lot" more than they would otherwise.
The author prevents and argues against a perverse strawman of the least likely results of there being a flourishing long tail of both content production and content consumption and tries to disprove it by focusing on the profit margins of middlemen and high volume producers.
I would argue that a healthy long tail will force the profit margins of middlemen down because they have more competition so there's less ability to extract from consumers. Same for high-volume producers - some of the market will shift to lower-volume competitors. And that's exactly what we see happening: incumbents have had to respond to markets moving away from them. A film or TV show has to compete with entire new worlds of types of video content, so those incumbents go broad. But that's because the long tail is thriving on its own, and doesn't need them, not because the production or consumption worlds are short-tail.
You can't monopolize and extract rent from the long tail the same way you used to from the short tail because the barriers of entry for producing long-tail culture products have gotten so much lower. The struggles of the former kings and incumbents shows that their relative market share is being eaten into by cheaper-to-produce alternatives that their business model simply can't sustain itself by competing with. If demand was short-tail then the biggest incumbent players would face far fewer barriers to raking in profit. Relative demand for short tail stuff goes up -> ability to charge a higher price for short tail stuff goes up. But that's not happening. (Disney is the closest thing we have to this, they have a pretty firm grip on tons of non-free family TV and film, but that's one of the few areas (along with live sports) that have been less vulnerable.)
Did I say "eliminate"?
The "long tail" concept however is not that "some smaller percentage would still buy less popular items if only they were made available". Even metal-polka-combo bands or noise bands making noise by hitting bricks together had some fans, so obviously if such less popular items were available of course somebody would buy them or stream them. That was trivially correct, and uninteresting.
The "disruption" was about how "products in low demand or that have a low sales volume can collectively build a better market share than their rivals, or exceed the relatively few current bestsellers and blockbusters, provided the store or distribution channel is large enough" (Wikipedia)
There are a lot of these markets that weren’t being served thirty, forty years ago, so it doesn’t seem so trivial to me to claim that business can exist to serve them. Nobody thought that at the time, given the tech or the time. Becoming cumulatively more popular? I don’t know where you draw the line, but I’d be surprised if the cumulative long tail of video content has not exceeded the viewership of Hollywood movie and TV products for the 30-and-other demographic. Same with Reddit vs major magazines.
The world is really big.
> Sometimes you will see a 90/10 relationship or a 75/25 split.
It's funny that his examples add up to a nice round 100, but it doesn't have to. The number of customers and percentage of revenue are two independent measures, and they don't have to sum to 100. You could have 20% of customers giving 99% of revenue (a 99/20 relationship) or 10% of customers giving 60% of revenue (a 60/10 split). I assume the author is well aware of this, but likes his examples to sound like the well-known 80/20.
Not to detract from your point at all — just a curious observation.
* In a way assuming continuity is weird because the underlying numbers are discrete. But in approximation it's fine here if the number of customers is large enough, I'd say.
Okay, you are probably not really interested in the point where they sum up to 100, and you just want the 80% of the profit, whether the distribution gives 80/10 or 80/78 (where all customers give you almost the same profit, revenue and cost). Still a useful rule.
I despise the “classic rock” format where you will hear the same tired song in the morning and evening drive time every day, there is a show on a local college station that plays “classic rock” B sides, obscure tracks, bands that sound like “classic rock” bands you know but did not get so famous. The DJ apologizes if he plays any track he played in the last six months.
Today my thing is anything that is recorded in a surround format be that traditional Chinese music or 2000s electronica or remnants from the 1970s quad era.
New music that appeals to the short tail has to get my attention through all that.
This is nothing. Back in Aichi Prefecture I had a neighborhood pirate DJ-cum-Yakuza errand boy who would cut off a finger every time his MD player skipped!
Youtube's search has become useless (to the point where it has become a meme that people are now using non-Youtube search engines to search for content on Youtube), and its recommendation algorithm is designed to promote what is popular (the short-tail).
It's still possible to create a new channel on Youtube that offers something different, of course. It's just harder to get visibility on that content because the same "chase the short-tail, ignore the long-tail" strategy is starting to find its way into how modern search engines behave as well.
I used to disagree with the claim that if you didn't rank in the top 5 search results you didn't rank at all. My disagreement was based on the fact that I earned a living from a website that received most of its traffic from organic search results. These days though, I think that's becoming more and more true based on how search engines present results to the user.
Yes the offer of music and video has increased substantially, but "long tail market"? Are there big enterprises making bank off niche consumers? I'm not so sure.
As a consumer I don't care much either though, as long as the content is there.
Here’s what I think they miss:
Netflix went from being part of the long tail to becoming an incumbent. Netflix then tries to find more big hits to maintain big incumbent winnings. And incumbents have been in many cases doubling down. But that’s exactly because the long tail is eating the “medium tail” that incumbents used to profit off of alive.
But that doesn’t mean you can’t use the long tail to bring in money. Though a competitive long tail will restrict just how high your margins can be.
The long tail threatens Netflix now (if only to an extent: something like Squid Game becoming a hit in the US would blow the mind of someone in 2006 and seem like a perfect example of long tail content being able to not just be economically viable but potentially turn into a huge winner).
But it threatens Spotify much less! They thrive off of it! The music world isn’t a bunch of services with exclusives trying to be winner take all, most content is across services.
Similarly it threatens Amazon, AliExpress, Etsy, etc, much less than Netflix.
The money in a “produce content for a small niche” was never going to be big for an individual producer, the money in that was going to be in being a successful aggregator, but in a competitive market you can only squeeze that so far.
But if the long tail was dead in terms of people being able to get exposure and views and listens then the YouTube and Spotifys of the world and the big labels would be raking in cash like they used to. They’d have a product without much competition and they could charge more for it from all sides, and take more off the top from artists too. Instead they’re in a pickle because it’s hard to compete with free and the free long-tail is getting bigger and bigger and bigger in terms of popular appeal (like things like youtube videos of people playing video games, which simply didn’t used to take ANY viewership away from TV shows).
I'm a tabletop roleplayer (have been one for over 4 decades). I have more content legally available to me now than I ever have before. I've recently bought (ebook) copies of obscure RPG material from the 70s that I've been looking for for years. I can easily find quality content for current RPGs (like D&D 5E) even if it's extremely nich, and simultaneously and I buy great new material explicitly written for for AD&D's First Edition!
The author makes several references to a consulting project where they also state that serving niches is not viable, this is probably true for whatever client was looking to branch into this. With the provided examples, the article reads as a defense of those previous findings.
The "long tail" concept about content, as made famous in the 00s by Chris Anderson's best-seller, is about the distribution of content consumption, not it's mere availability.
Basically about how many views/streams go to a small ("popular") part of the catalogue, and how many to the long tail.
Having more content available (or some people watching less popular stuff) is required but not sufficient property to have a long tail effect.
The Long Tail is a way of saying: "There's a ton of tiny markets out there that you can go and serve, and make a decent, if small, living off of."
That's your shot at your own little 80/20 microcosm, if you do it right.
Yeah everyone knows Call of Duty, they sell like 20 million copies right?
But then there is a game series called Atelier which has been around since the late 90s and they sell to a dedicated audience of sometimes literally less than 100k. And these games are expensive.
https://store.steampowered.com/bundle/9121/Atelier_Arland_se...
Turn on FM radio, surf through the low number channels on tv ... the difference between that and what I consume on the net has a venn diagram of two distant far away circles.
Not only, but the concentration of the top tier has diluted. Just look at the rating for the most watched shows: https://en.wikipedia.org/wiki/List_of_most_watched_televisio...
If you exclude sports, the ratings are half of what they were pre-broadband.
There used to be a collective culture dictated by presumed media consumption up until maybe 15 years ago. That's also gone.
The long tail has been so victorious, it's overtaken the dominant framing so it can no longer be seen as distinct. It'd be like saying "jet airplanes" were no longer a thing because we've dropped the term "jet" when in reality, they've become so dominant for commercial air-travel, the comparison for the normal passenger no longer exists.
The core long tail claim, as I remember it, was that the _area_ under the long tail was larger than the _area_ under the head.
Sometimes much larger.
I vaguely recall a claim that many businesses chopped off the market tail because it was often too expensive to serve those tiny segments (inventory, retail space, distribution, etc.), but that new technologies were reducing or eliminating some of these costs and making it possible to serve emerging, smaller markets profitably.
I wonder if some portion of businesses that see pareto distributions in their sales are just lopping off long tail revenues? I thought that was the central claim.
I also wonder if those pareto variants (i.e., 95/5 splits) were businesses that had potentially over optimized on their largest volume offerings?
Amazon was the quintessential long tail example, and I'd be surprised if anyone could claim that their their SKU count has decreased since 2006.
YouTube comes to mind as another great example of a business capitalizing on this phenomenon.
Consolidating markets (like ??) may be counter examples.
Now I'm curious about how to choose when to push in which direction.
But the nature of the long tail is that it is comprised of a series of outliers. To use the author's analogy, what might be happening is that there's a distributional shift occurring at the same time the general amount of "stuff" under the curve is increasing, which gives the appearance that things have become more alike, even though what has actually happened is that the bump has just gotten bigger.
For example: https://onlinepubs.trb.org/Onlinepubs/state-of-the-art/2/2.p...
Of course these aren’t easy to find, and it’s only through specific searches that a hidden away scanned pdf put through OCR can be found. It would be great if there was more of a movement to scan all this material and put it online.
One example from the days of the origin of the Long Tail was Bollywood movies.
There aren't enough fans of Bollywood in one geographic location to warrant dedicating a whole theater to a showing. (I'm ignoring regional pockets, of course).
However, if a movie gets lot of views online, that might boost it up the tail into the attention of more casual fans. In turn, they might boost-like it into more mainstream etc.
That was much more difficult in the past where even if there was a lot of word of mouth praise, you still had the geographic distances.
Where did the long tail go? - https://news.ycombinator.com/item?id=31883206 - June 2022 (208 comments)
To start with, human society is very prone to long tails by the fact we are distributed widely and have limited means of communicating and coordinating.
All sorts of inventions, technical and social work on this substratum and produce more or less long lived tail phenomena.
E.g. the internet famously enables the long tail, but then the first page of google effect squashes it. Similarly social media platforms enable cheap and easily accessible dissemination but then algorithmic timelines and virality enhancers intervene to create "winners"
There is a complex interplay between concentration of power, wealth, control and technological capabilities on one side (the left end of the distribution) and the shape of material and information flows on the other (right end of the distribution or long tail).
I was going to write a counter-counter-culture bit eulogizing that shared media landscape, but the fact is media is not culture. Culture has existed long before media, and is much larger than that. Shallow similarities like watching the same shows cannot replace loving one another.
Recently I learned that my teenage son purchased a movie called “Dad’s Can’t Dance”, which is only available from this website: https://www.daddycantdance.com/
Why? I have no idea (why do teenage boys do any of the things they do?), but there can’t be many people buying this movie off a random website.
I sometimes wish we could go back in time to when you could only reach people by landline and there were only two TV channels.
Step 2: ????
Step 3: Profit.
For example, all the indie music is on Soundcloud and Bankcamp, not on Spotify.
The first is the idea that, as a marketplace for goods, there are a lot of--maybe (if you go all in on the premise) even the majority of--sales to be made within the long tail. Maybe you believe the 80/20 split the author here posits is a good reason to discard the 20%... I think that's dumb and I think that both Amazon and Spotify are constantly cited somehow in this article as doing something wrong demonstrates as much.
But... the article spends most of its time moralizing about a very very different idea: a frustrating claim that anyone who believes in the long tail ever promised that if someone allocates their effort to making a product for a niche audience that they will make a bunch of money because "the blockbuster is in decline" not merely in total sales volume (due to the relatively new accessibility of the long tail) but is somehow a worse bet than making a niche product for a small audience.
That is not only not the case in practice it is actually trivially not the case from the very idea of the long tail: every product on the long tail isn't making many sales by definition of where they are in the distribution. At one point this author finally states as much... but somehow it is against the long tail people, as this idea that people should go out of their way to make content for this segment was attributed to them as the most disingenuous form of strawman throughout the rest of the article :/.
The long tail argument is merely that if you sell a bunch of stuff that a non-negligible amount of your sales comes from the long tail, not that the long tail is somehow itself fat and that the people making those products are going to ever manage to pull off a bunch of sales. If it is wrong, it is because people might want it to be a majority and not just "a lot", but FWIW in my experience 20% is "worth it", and if I went to a supermarket that only stocked the 10 most popular foods I'd find a better supermarket even if 80% of my purchases and 80% of their sales are from those 10 foods: it is really hard to run a business that only serves blockbusters.
But... is it sane to allocate a bunch of time and money to making something that isn't a blockbuster if your goal is to maximize profit? Of course not, and no long tail advocate ever claimed as much! If you're concerned about where all that content comes from--and I can tell you this as someone who ran a very popular and even culturally-relevant content marketplace--it is largely people working on low or even negative budgets and for whom even a single sale is a life-altering moment.
But like, if you take a percent of that person's one sale and you add it to the numerous other peoples' one sales, does it add up to "a lot" of money? Without even having to chime in with my own experience, we can actually note that even this author seems to claim--by their assertion in 80/20 or (later) 75/25--you could increase your revenue by 25-33% by bothering to open up your catalog... and that's even more the "I don't bother to shop at a supermarket where I can only find 90% of what I want"!
The author thereby needs to stretch pretty far to explain away why markets like Spotify and Amazon disagree with the article, claiming that Spotify is I guess just wrong and will likely change their strategy in the future once they realize this author is smarter than them and that Amazon makes most of their money on AWS and so (supposedly) isn't a relevant data point... maybe it would surprise the author, then, to learn that Apple iTunes also allows pretty much everyone to list their tracks (and thereby contains an extremely long tail that certainly doesn't make people a lot of money but which absolutely helps cement them as a go-to destination to buy and collect music) and that, while Walmart has to make some tough decisions on shelf space, they have an extremely expansive online-only catalog.
The only example they have of a marketplace (as the idea that a movie producer concentrates on blockbusters is not relevant to the premise) shrinking their content catalog is Netflix, and that's not at all Netflix's conscious strategic decision: they only were able to have all of that content for so long because the people who owned it didn't want to directly compete with Netflix and were willing to license it "on the cheap"... now that they all want their own streaming service, they are clawing back their interests region by region, and it frankly isn't clear that the gutted separate services are going to manage to sustain their users as well as the unified market could (which I will claim is in no small part due to long tail effects; their best strategy will likely be to re-form the old cable market by joining forces under a single subscription).
I'm honestly not sure why this article has managed to captivate so many people given how mixed up it is in the premise, but I hope that readers spend some time really thinking about not only whether the points made in this article resonate in the order presented but whether they manage to correctly lead to the moral and economic conclusions they purport to before anyone ascribes to "the long tail" something only this author incorrectly understood and start--as I have seen many do--using it to justify actively shrinking a catalog or the idea that a large catalog isn't somehow a powerful moat (which someone tried to use against me once to claim how easy they thought it would be for just anyone to compete with Spotify).