Paywall

A paywall is placed, and where it is placed is most of the design. A hard paywall sits in front of everything: the headline, maybe a deck, then the offer. A metered paywall lets a counted number of articles through and appears when the count runs out, which is why it carries a number the reader can check (“2 of 3 free articles this month”) and a hard wall does not. Between them sit the freemium wall, where some pieces are free forever and others never are, and the registration wall, which takes an email rather than a payment and is a different trade dressed in the same furniture.

The moment of the wall is a moment of loss, so the craft is in not making it feel like a trick. Show the counter before it runs out, not only after, or the reader experiences the meter as a trapdoor. Say what the subscription costs on the wall itself rather than one click later. Keep the sign-in route visible and equal in weight to the subscribe button, because a large share of the people who hit a wall are already paying and are simply logged out on a new device. And let the reader see enough to judge whether the piece is worth buying: a headline over a fade is usually not enough, and a fade over three paragraphs usually is.

The fade itself is a specific bit of vocabulary. Clipping the text and washing it into the background says “there is more” in a way a hard cut does not, which is why almost every metered wall uses one. It also invites a cheap mistake: rendering the whole article and hiding it with CSS, so anyone who opens the inspector reads it free. If the content is meant to be paid for, the server has to withhold it, and what the fade covers has to be a genuinely truncated excerpt. The related trick of serving the full text to search engines and the wall to people arriving from them has a name, cloaking, and search engines penalize it; the sanctioned version is a structured-data flag that declares which part is paid.

Two things separate a paywall from the dark patterns it sits next to. It should not manufacture urgency it does not have, so a countdown on a subscription offer that resets on reload is lying. And it should not make cancelling harder than subscribing: if signing up is two clicks in the app and cancelling is a phone call during office hours, the pattern has stopped being a wall and started being a trap.