I came across an interesting example of something that has been normal in airlines, hotels and e-commerce for years, but still feels slightly disturbing when applied to journalism:
personalized subscription pricing.
Nieman Lab reports that publishers including Wired, NJ.com and The Wall Street Journal have been using algorithms to determine renewal prices for individual subscribers based on personal data and behavior. Nieman Lab
The screenshot above is a particularly nice example. Wired has a regular annual rate of $80, but the subscriber received a personalized renewal offer below that level. Other reported cases are even more illustrative: NJ.com subscribers comparing notes discovered annual renewal prices of $130, $145 and $175. Nieman Lab
Same publication. Same product. Different human. Different price.
Welcome to SaaS, except you’re buying journalism.
“This Price Was Set by an Algorithm”
The reason we’re suddenly seeing this explicitly is regulation.
New York’s Algorithmic Pricing Disclosure Act took effect this year and requires businesses to disclose when personalized data was used to determine a price. Nieman Lab
Hence the wonderfully dystopian sentence appearing in renewal emails:
THIS PRICE WAS SET BY AN ALGORITHM USING YOUR PERSONAL DATA.
Which is refreshingly honest.
The interesting question for me isn’t whether publishers will continue doing this. They probably will.
I’m more interested in which creator platform will be brave enough to do it first.
Substack Is Already Halfway There
Substack doesn’t currently assign every reader an individually optimized subscription price. But it already has something surprisingly close to the machinery required.
Substack Boost uses engagement data to decide when particular readers should see trials, discounts and other special offers. It can score free subscribers by likelihood to convert, automatically offer discounts, and intervene when paid subscribers appear likely to churn. Substack Support That’s not personalized pricing in the strict sense.
The base subscription still has a publisher-defined price. But conceptually we’re already here:
Reader A → full price
Reader B → 20% discount
Reader C → free trial
Reader D → retention offer
The algorithm isn’t setting the sticker price yet. It’s deciding who should actually have to pay it. The distinction is getting rather thin.
And Newsletter Platforms Are Becoming CRMs
The other half of this story is what is happening at Kit. Kit recently launched Subscriber Signals, which enriches newsletter subscribers with professional, social and demographic context.
Instead of seeing:
john@example.com
opened 14 emails
clicked 3 links
a creator can increasingly understand things like: role, company, location, social profiles and reach, age and demographic characteristics, purchase history, engagement, and potentially whether this person looks commercially interesting.
Kit says the system can enrich roughly 25–30% of a typical list, sometimes considerably more, using third-party data providers. Kit Help Center And it isn’t merely analytics.
Creators can search for CEOs, identify influential subscribers, build sponsor reports, segment audiences by industry or demographics, and target offers accordingly. Kit This is a meaningful change in how newsletters are conceptualized.
A Subscriber List Is Becoming an Asset Database
For years, newsletter analytics mostly looked like this:
47,000 subscribers
42% open rate
3.8% CTR
Useful. But economically primitive.
Imagine saying that a company has 47,000 customers and then refusing to learn anything else about them.
What Kit and similar tools are moving toward is much closer to CRM logic:
Who are these people?
Where do they work?
What can they afford?
What are they interested in?
Who influences others?
Who buys?
Who is likely to buy?
Who might become a sponsor, customer, partner or affiliate?
And once you start thinking about audiences this way, personalized pricing becomes almost inevitable.
The Creator-Economy Version Gets Interesting
Imagine a newsletter with a nominal annual price of $100. The platform knows Reader A is highly engaged, has read you for three years and has previously bought two products. Reader B subscribed yesterday and barely opens anything. Reader C is about to cancel. Reader D is a senior executive at a Fortune 500 company. Reader E lives somewhere where $100 represents considerably more purchasing power.
Today we mostly charge all five the same price and occasionally throw discounts at them. From a revenue-optimization perspective, that’s rather crude. An algorithm could theoretically optimize:
price × conversion probability × retention probability × lifetime value
for every subscriber individually.
The newsletter suddenly stops having a price. It has a pricing function. And this is where creator platforms start looking much less like blogging tools and much more like miniature commercial operating systems.
From Audience to Portfolio
I think this connects to the broader direction we’re seeing in creator businesses. The valuable thing isn’t simply having 100,000 subscribers. It’s knowing what those 100,000 people represent.
An audience can contain customers, executives, investors, sponsors, influencers, future employees, corporate buyers and people who will never spend a cent but will distribute your work to thousands of others.
Treating all of them as identical email addresses with an average CTR increasingly makes no sense. This is also close to the logic behind the emerging audience-as-asset model we’re seeing around businesses such as Workweek: the audience isn’t merely distribution for content. It’s an economic asset that can be understood, segmented and monetized in multiple ways. And AI makes the economics of doing this much easier.
A traditional CRM with 100,000 individual records requires rules, segmentation and people operating it. An AI-native system can theoretically reason about 100,000 individual relationships. That’s a very different thing.
So Who Does It First?
My guess is that personalized creator pricing will initially arrive disguised as something friendlier.
Not: “Anton pays $87 because our algorithm thinks Anton will tolerate $87.”
Probably: “Smart offers.” “AI revenue optimization.” “Dynamic retention.” “Personalized membership.” Something with a pleasant gradient in the dashboard.
Substack Boost is already an early version of that idea. Kit is building increasingly sophisticated subscriber intelligence around the other side of the equation.
Eventually somebody will connect the two. And then your favorite newsletter may no longer cost $100 per year.
It will cost:
whatever the algorithm thinks you personally are worth charging.
At least New Yorkers will get a warning.