Everything I have ever learned about subscriptions and ads.
Also calendars and browsers.
Chris Hua’s perpetual motion machine of hot takes.
Power-knowledge.
My general thesis on creator economies are that they’re hit-driven. For writers, you are better off writing for a newspaper or magazine (if you can!) if you are not a big name. When you have name recognition and a following, starting your own publication (like a Substack) makes sense. The publication offers you distribution, which is so valuable. The tradeoff is that you don’t own the relationship with your readers. Both of these are sources of value capture, but it’s very hard to have both.
In the past, the face of “content” was organizations, rather than individuals - we personified the New York Times as “the Gray Lady,” not “a platform to reach the individual journalists we love.” In those times, you could argue that the NYT, as an entity, had both distribution and owned the reader relationship. They made money on ads, they made money on subscriptions, life was good.
One of my core theses about life in the 21st century is that when everything is connected, nothing is connected, and every single relationship becomes atomized. There will never be another New York Times. More on that in the future…
Twitch and Luminary are good examples of two of the main types of subscription pricing models: Twitch allows you to subscribe to individual streamers, and Luminary allows you to subscribe to a platform with many podcasters. NB: Yes, Twitch has a “platform subscription” too, which removes most ads from the platform, but I don’t think it’s a material amount of their revenue (very rarely see the Turbo sign in chat, not quantified, sorry!). In particular, if you subscribe to a few top streamers, and spend most of your time watching those streamers, you won’t see ads most of the time anyways.
Now both of those platforms are struggling or starting to suffer.
When consumers subscribe to individual streamers, the streamer owns the consumer relationship. When the streamer leaves, say, enticed away by a big money exclusive deal elsewhere, so go those consumers. That’s what Twitch is running into as the market for (buying) streamers becomes more competitive and efficient. Twitch still has the best distribution, i.e. most viewers and probably acceptable recommendation algorithms, which matters for smaller or mid-tier streamers, but isn’t that important for the very top streamers. And in a hits-driven business, that isn’t enough.
When consumers subscribe to a platform, the platform owns the consumer relationship. Luminary is an example of that, and creators really didn’t like their practices:
Podcasters also pushed back on Luminary’s practice of stripping podcasts’ show notes, which sometimes have sponsored links and other information, and of the way Luminary rerouted listeners so that podcasters couldn’t tell where traffic was coming from.
Numerous high-profile podcasts pulled their content from Luminary after its launch, including shows like The Daily, The Joe Rogan Experience, and Blackout. Major publishers like Gimlet Media and Parcast were never on the platform to begin with, and PodcastOne and iHeartRadio removed their networks of podcasts from distribution on Luminary.
But you see why Luminary makes these choices? They need to aggregate listener demand and own the relationship between the user and the podcaster to capture value. Under aggregation theory, if you aggregate demand then suppliers have to come to you and operate under your terms.
It didn’t work. They’re lowering their prices significantly and adding a yearly subscription discount.
Some thoughts on why it didn’t play out:
Too much free content and too many listeners - the demand and supply sides of the podcast market are both very fragmented. Obviously you should put that in your VC pitch deck, but a good VC also should not take that line as a reason to invest. Baudrillard had this great line, “the word is free, but I am not; the [airwaves are] so saturated the pressure of all which wants to be heard so strong that I am no longer capable of knowing what I want.” It’s hard to enter totally fragmented markets!
People listen to podcasts to be entertained or informed, sure; but that action cannot be divorced from the connection which the listener shares with the creator. The barrier to entry to podcasting is low (see also the first point about too much free content) so it’s easy to perceive a personal connection with the host. The good hosts are playing heavily into this connection, e.g. through merch sales, live shows on tours, etc.
I would not make this argument for platforms like Netflix, where most watchers are so sufficiently different from the people on the other side of the screen that there is no real personal connection. I like watching David Chang, but I’m not under the illusion that “flying to Morocco with Chrissy Teigen” is in the cards for me anytime soon.
The biggest podcasters have already built that listener relationship; if you want to listen to Joe Rogan, you’ll find an app that lets you listen to him. And hey - the built in apps on iOS and Android work just fine for finding his podcast. So, podcasting is a hits-driven business as well and
Trevor Noah and Lena Dunham aren’t funny and you couldn’t pay people to listen them.
Advertiser atrophy.
There’s a post about new startup ideas that got a lot of internet points on news.yc. Yay.
Ad companies are governed by a rule as durable as gravity: they must make their products worse over time. Corporations must produce profit, and the fastest path to increase revenue is at the cost of user experience.
So I generally agree with this statement, but would say that the consumer products get worse over time. The big advertising products are great! Advertising is truly a business which gets better at scale - more advertisers, more users, better targeting algorithms all directly translate to more revenue. The prospects for new ad-supported entrants into the social space are abysmal. Even if you have a great product, engaged users, and perfect machine learning algorithms, if you don’t have scale, you can’t convince advertisers to advertise with you. That feedback loop both helps the big boys and cripples everyone else.
TikTok might be the one exception here, but what I have heard is that their advertising is still focused on large brand buys with specifically tailored creative. A large reason behind why Facebook and Google dominate internet advertising is because they enable the long-tail of advertisers to get started and produce ROI with little friction. Similarly, Reddit and Twitter are taking brand-focused rather than performance-focused approaches to advertising. These are rational decisions given that they cannot reasonably compete with F/G for performance dollars outside of certain niches, and they have lots of inventory they need to fill.
Anyways, here’s another line that I may take out of context to harp on a bit:
finding a genuine phone charger on Amazon requires a degree in investigative journalism.
Again, I agree with the statement, but I think the guy misses the point. Yes, Amazon ads are really terrible and pervasive (at one point, I did an audit and estimated Amazon showed 2x the ads of competitors), but the core reason the Amazon experience is bad is because there are too many products. It turns out finally building “the everything store” means that everything is on the shelves, and nobody can find the right thing. That’s a core part of my thesis behind why Square, Shopify, Stripe, and other fintech companies starting with the letter ‘S’ will dominate the next decade — people want a choice of where to shop, and a curated selection of things from those choices.
Retail isn’t getting better with scale but advertising is — to their respective customers.
AI powered calendars.
This isn’t really a “creator economy” story but one thesis behind the creator economy is that people are willing to spend money to support things they love and to have a better experience. Superhuman is the poster child of the latter category, $30/month to have a slicker, faster email experience. It is also currently blocked on my corporate email and I’m slowly weaning myself off of it….
Anyways Clockwise is a really compelling product in theory - use ML to schedule meetings and protect free time. I tried it at my last company and had some thoughts:
The extension was pretty buggy and slowed my GCal window down a lot. I preferred using the mobile app over desktop…
It only really works if everyone on your team uses it. Maybe that’s their lock-in/moat, and suggests that they should pitch it to PM’s or CEO’s and get their entire company on board. But if you think of how Dropbox grew, the early adopter who pushes the rest of the company to use it, I don’t see the same kind of value proposition.
It was very good at getting fewer meetings. It puts these huge “focus time blocks” on your calendar and then people are too scared to ask you to meet. So you don’t end up having any new meetings. Obviously there are pro’s and con’s here.
FiredFox.
“You may recall that we expected to be earning revenue in 2019 and 2020 from new subscription products as well as higher revenue from sources outside of search. This did not happen,” Baker writes in her memo. “Our 2019 plan underestimated how long it would take to build and ship new, revenue-generating products. Given that, and all we learned in 2019 about the pace of innovation, we decided to take a more conservative approach to projecting our revenue for 2020. We also agreed to a principle of living within our means, of not spending more than we earn for the foreseeable future.”
Subscription products trade off short-term revenue for greater long-term recurring revenue. From a discounted cash flow perspective, that’s fine — yeah in the short term you have tighter income, but you’ll make up for it with greater income in the future. From a personal perspective, well, you might be out of a job.
I guess a couple thoughts:
Why did Mozilla have so many people and such a nice office in the first place? The counterfactual would be projects like the Apache Project, which are distributed by nature and subside on much smaller amounts of funding.
Why didn’t they bridge the gap between short-and-long term funding with a loan?