Venkatesh Rao Profile picture
17 Nov, 40 tweets, 8 min read
bluecheck invasion ruining substack a bit 😬
I have more hope for the platform than for Medium or Patreon (that's why I'm on it). It has less of a top-down cultural feel, and little to no cross-talk/cross-flow across newsletters, which is a good thing. Sacrifices content-network effect for diversity.
Basically the bias-variance tradeoff writ large. An aggregator platform will either have a large bias or a large variance. So far, unlike Medium or patreon (or livejournal or tumblr or what have you) substack is heading towards higher variance. x
But the arrival of 800lb gorilla bluechecks like Greenwald and Yglesias means that there is going to be an optics problem. They'll account for 1% of the writers, 30% of the unique subscribers, and 80% of optics. So even if the base has high variance, it will look like high bias.
The optics will drive preferential attachment of both readers and writers. And increasing number of readers will be singletons attached to single bluecheck-grade writers. Which means the exit-risk cost of a these single writers will be disproportionately high.
Unfortunately, the incentives are going to be tough for @SubstackInc because the exit-risk cost associated with teh 800lb gorillas means there will be pressure to overserve them, and underserve the long tail. The long tail is lower-margin.
@SubstackInc I've chatted with Chris and Hamish. Great guys, and the fundamental sanity they seem to be bringing to this is a big part of the reason I got on board in April 2019, about a year before the bluechecks began discovering it. But their job is going to be really tough here very soon.
@SubstackInc To lower singleton-subscriber risk associated with big bluechecks (ie readers only subscribed to say Greenwald), the most obvious option is to drive cross-promotion. Ie, attach their readers to other writers too, so platform revenue isn't sensitively dependent on 800lbers.
But the *moment* you decide to exercise the cross-promotion option, you're no longer editorially neutral. You're moving away from variance and towards bias on the bias-tvariance tradeoff curve.
The thing is, the easiest cross-promotion to do is the kind that reinforces existing tribes. A "if you like this, you'll like this also" approach will hook greenwald subscribers up to writers who are also like greenwald. Going against grain will be hard.
Along-the-grain cross-promotion will lower fragility a bit wrt single 800lb bluechekerillas, but simply bump the fragility up a level to an existing tribe. Patreon discovered this the hard way. Now you have to scenario-plan the exit risk revenue hit of entire tribes.
Worse, the tribes will realize it too, and will achieve a platform-local tribal consciousness and begin trying to "own" the platform, regardless of whatever the cross-promotion algorithms do along/against the grain. The platform will be drawn into a fight against its own readers.
I think Substack is doing many of the right things for the current stage of the platform. For example, focusing hard on empowering writers rather than readers, and doing a lot for beginning writers trying to build an audience... that's all good.
Reshape the curve away from 800lb bluecheckerrila dependence. Ideal for people like me, the mid-level non-brand-name types would be if substack could afford to lose any single bluecheck and never have to go to cross-promotion at all. This requires healthy margins from long tail
This is why it is short-sighted to complain about their high fee percentage (10% on top of credit card fees as opposed to say the fixed fee of ghost or mailchimp). They need to be making enough money off the long tail to resist the pressures created by 800lbers.
I don't grudge them that. I'm open to things like ghost and memberful, but it is polyannish to believe they resolve the issues. I was on mailchimp for a long time (still am for my basic ribbonfarm list) but their incentives for eg. push the product towards marketing clients.
Ghost is new enough that it isn't showing its own vulnerabilities, but there will be some. No free lunch. Content platform Pick 2 of 3 triangle: optimal featureset for you, low cost, high variance/low-bias.
An instructive precedent is the adjacent courseware sector. The el cheapo platforms really don't work for indie course creators. Teachable does, BUT costs an arm and a leg and STILL had to sell itself to a Brazilian PE firm to survive.
Courses require way more work to turn into profitable low-touch income streams. My teachable account just barely pays for itself because I don't put in the work to keep publishing new courses, running live cohorts, or updating old courses.
I'm considering EITHER kicking my teachable stuff up a notch next year or shutting it down entirely. I'm in an untenable place with it long term. Kicking it up a notch will require periodic effort spikes, as opposed to newsletters which require steady effort commitments.
But to return to substack. I a) like the product b) am willing to continue paying the price premium to keep it long-tail/variance oriented c) don't like the alternative incentive pressures competing platforms face.
But one thing I'm not doing is moving home base from blogging like I was considering last year. Wordpress is an aging monster in many ways, but it is still the worthwhile foundational piece of the puzzle.
The one wildcard in this picture is where static sites will go -- gatsby, jekyll etc. I've now been on the sidelines of one pretty ambitious static site ( and developed a better appreciation of the potential there, but definitely not for pure writers
Still rounding out the views I put forth in my text renaissance post earlier this year, and might potentially use the ideas to expand the longform course on teachable so I can make money off all this content industry prognistication…
This is my art of longform course. I need about 2-3 weeks uninterrupted time to record a 2x expanded version covering all this new stuff, but haven't been able to find the time. I don't think I'll have the time to run a live cohort, which would be ideal.…
This is a good point. WordPress violates my platform point by... simply not being a platform at all. It is featureset optimal, can be low cost (if you self host and have low traffic) and high variance.
I think this is because Automattic failed at making wordpress dot com a true platform level competitor to things like Medium. It never outgrew roots as basically an open source product rather than a platform. What it actually competes with are hosted WP players like WPengine
I've been on WPEngine a long time, happily paying premium to not have to think about the headaches of self-admin'ed WP. But hosted WP is not a platform and can never be. The tech is fundamentally a digital mansion. At most you can create a feudal landscape of mansions. Not a city
This article has apparently been doing the rounds ht @noUpside…
Will read and comment later on this thread. Stopping for some Real Work™
Ok, read it. Kinda meh. The writer kinda doesn't get it I think. The choice to center the kind of writer who is hoping to go to a secure paid gig at the same level of effort, lower risk, and more pay, is not really the interesting kind of writer for the future.
Like many rearview mirror analysis, the writer is looking at this through the lens of old media and the unexamined assumption that the same people must succeed on the new medium, for the same kinds of value/recognition. Begging the question basically.
If you assume the status hierarchies of the old world must self-evidently be sustained in the new world, of course you'll reach the conclusion that that is what will happen. The problem is there may soon be no "traditional" world for people "passing through" substack to aspire to
If you've been substacking with the hope of getting a NYT op-ed gig... you are kinda missing the whole point. The point is that utopian destination is increasingly not a tenable one at all and its own top people are abandoning ship. So this sort of thing is "it."
Also, I'd like to know how many fires I have on their 1-4 fire scale. I'm guessing a 1-fire.
My unsolicited top-3 advice items:

1. Stay the course focusing on writers over readers

2. Avoid cross-promotion

3. Build more individual writer branding tools and push tech into the background. Avoid temptation of making the platform the brand.
Basically the product would benefit from being more product-driven. True north is not NYT but WordPress. Deeper, more feature-ful product, more customizable for all. Definitely do NOT create 2 classes of writers (rating system is fine, different feature sets starts to get dicey)
tldr of strategy recommendation for substack: commoditize your complement, ie writers to the tune of 90% but not 100%. Putting all writer brands on an equal level playing field instead of creating *feature* tiers for bluechecks is the key to being a good platform.
There is no royal road to geometry

There is no special iPhone for special people, everybody gets the same price points.

No special substack for special writers. But fee tiers for ++ services are okay. Like the extra fee for own domain mapping.
This is possibly a problem for history-completist readers, but is frankly not a big deal for me. I'd be fine with most of my content just vanishing into random blackholes so long as I've gotten what I want out of creating it.

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Venkatesh Rao

Venkatesh Rao Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @vgr

17 Nov
Anyone actually enjoy the zoom-conference circuit? I’ve pretty much entirely avoided it. I’ve realized that to the limited extent I enjoy conferences at all, it’s 100% about getting to travel to places I want to visit. Take that out and it’s just feels like work to me.
The one thing I enjoyed was a very lowkey set of maker-skill demos hosted by a local repair cafe.
Looking back on a couple of decades of conferences, the memorable things I actually remember are the local side trips.
Read 4 tweets
17 Nov
There’s an episode of Star Trek: Voyager (S5E1, “Night”) where they’re crossing a vast void with no stars visible because black clouds shroud starlight. The crew is going crazy. Then stuff happens, but it’s the set up that interests me: featureless darkness as a psychic stressor.
It’s a nice allegory for why a dark age might be stressful even if things are otherwise fine. The lack of macro patterns makes microstructure of life start to empty of meaning and purpose. The Voyager doesn’t need visible stars to navigate but the crew needs them for sanity.
Ancient cultures made up constellation maps and astrology out of otherwise mostly useless night sky landmarks to create macro patterns I suspect. Limited navigation and calendars needs don’t explain the overfitting of mythologies to the skies.
Read 17 tweets
17 Nov
I’m getting good enough at spotting very precise tropes now so I spoil tv shows for myself. 😖
The one that inspired this tweet: a passenger left alone in a car while driver goes into a store will discover a secret in the glove compartment. It’s not really a trope but a shot clue: if they show the left behind person rather than follow the other character there’s a reason.
An annoying one: in a detective show, a recognizable character actor showing up in an early scene in a minor role with unnecessary lines is probably the murderer. You don’t need to know the plot to guess that they wouldn’t use a speaking role established actor as an extra.
Read 4 tweets
17 Nov
Ok I’m not starting any more projects. At least not top-level projects. From now on I’m only finishing projects or killing them.
My biggest achievement of the last couple of years was effectively ending Refactor Camp the conference on a graceful high note. Second biggest was handing off the refactorcamp dot org the mastodon server to someone more able to keep it going
Feeling very post-mid-lifey, leeward sloping. Turned 46 last week. Maybe act 1 is starting all possible projects, act 2 is shutting them down at the natural times.
Read 8 tweets
16 Nov
“It is 2020. 250k are dead from a raging pandemic; a private space company is launching crewed missions NASA can’t. A mad president is live-tweeting a painfully slow psychotic break, refusing to leave office. The planet boils.” —Blurb from “2020” obscure sci-fi book from 2007.
I seriously couldn’t have imagined this plot in 2007
Actually I can’t imagine it right now either, even though it’s literally happening all around.
Read 6 tweets
15 Nov
brrrcore: the thing that goes brrr, assuming it exists, when you remove deadening constraints, complexity, ornamentation, and superfluity from a thing, and distill the rest to it’s essence. The id or kundalini of the thing. For eg, the tweet is the brrrcore of the blogpost.
brrrcores are pure and unstable. Unstable positive feedback loops at the heart of anything that shows signs of exponential aliveness. The (auto)poiesis principle animating it. Getting to the brrrcore is the essence of any alchemical pursuit. Eg Network effect in product adoption.
Once the brrrcore is uncovered and allowed to energize itself, you can gradually add back the “impurities” you threw away while looking for it.
Read 6 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!