How Piano created a “propensity” paywall that uses machine learning with 76 metrics to predict how likely someone is to subscribe, and adjusts accordingly
They went from being the kid nobody wanted to talk with to one of the cool kids on the block. (Apparently showing the other kids how to make money helps.)
Context & Ripple Effects
Earlier in 2019, Piano raised a $22M Series B on a pitch of helping its 1,300 media-brand customers bring in non-ad revenue. The propensity paywall is the product answer to that pitch: rather than a flat metered wall applied identically to every reader, Piano scores each visitor against 76 signals and adjusts access based on predicted subscription likelihood.
That positioning is what turned Piano from a niche tool vendor into a core monetization partner for publishers — a trajectory the market confirmed two years later with an $88M Series B that brought total funding past $241M.
First-order effects
- Publishers running Piano can stop treating all anonymous readers alike: high-propensity visitors get paywalled sooner, low-propensity ones keep reading, trading short-term pageviews for modeled conversion lift.
Second-order effects
- Rival subscription and analytics vendors must match machine-learned paywalls or cede the upsell conversation, and pricing power shifts toward whoever accumulates the cross-publisher behavioral data that trains these models.
Third-order effects
- If dynamic, prediction-based access becomes table stakes, the paywall stops being a fixed editorial policy and becomes an algorithmic pricing surface — a structural shift that also primes publishers to price other forms of access, like the AI licensing deals OpenAI and ProRata are now negotiating with media companies.
The trend: Publisher monetization is moving from uniform meters and ad dependence toward algorithmically priced access, a shift now extending from reader subscriptions to AI-content licensing.