Let’s talk about Paywalls

Hub Uy
4 min readApr 4, 2022


Optimizing Washington Post’s Paywall

The Washiongton Post’s New Paywall


When non-subscribers visit The Washington Post, they are given 5 articles before being hit by a paywall.

The paywall can oftentimes be an annoyance (as the majority of users believe that news should be free). We want to therefore ensure that the paywall communicates the right information and the journey to conversion be as seamless and frictionless as possible.

Control User Flow

A Diagram of The Washington Post’s Userflow from the Paywall to Payment Confiriation
A Diagram of The Washington Post’s Userflow from the Paywall to Payment Confirmation


Our conversion funnel is skinny.

Only 2% of users that are hit by the paywall make it to the purchase consideration (Product Page). And only 0.1% make it to purchase.

The Washington Post Conversion funnel

How can we design a paywall that would help users in their decision-making process and improve conversions?

North Star

All purchase considerations should be made outside of checkout, allowing users to focus on 2 separate tasks — shopping and paying.

Express checkout is already an established user habit.

Applying Past Learnings:

❌ — Big pricing does not work for our users.

✅ — However, pricing on CTA button works!

✅ —Strike-through pricing works to communicate savings.

✅ — Toggles allowed users to easily compare options and increased conversions on more premium and long-term products (diversifying product mix).

🏆— 60% of users prefer express checkout options.

Test Variant 1


Users will have an easier time comparing offers if we showcase monthly and yearly offers on the paywall

Design for Test Variant 1 — Giving users options on the paywall to bypass the product page
Design for Test Variant 1 — Giving users options on the paywall to bypass the product page

Test Variant 2


Users will be more inclined to consider a purchase from the paywall if we expose express payment options to make payments easier

Design for Test Variant 2- Giving users an express payment option
Design for Test Variant 2— Giving users an express payment option

A/B Test Results

The test started on April 29, 2021. We reached statistical significance in 5 days but decided to run the test for 1 more week to gather more data. The test results below are from the data gathered as of Sunday, May 9, 2021.

🏆 Variant 1 increased conversions by 15.75% while Variant 2 increased conversions by 2.68%

A/B Test Results
A/B Test Results

Key Insights

  1. Surfacing express checkout allowed us to help users with breezing through checkout. In control, 50% of users would select an express payment method even without surfacing the options on the paywall. With the new designs, 60% of users selected an express payment option: +10% shift in PayPal purchases.
  2. Users want to make sure that they have explored and exhausted all purchase options. This is apparent in Variant 1, in spite of having the toggle as a point of friction, performed significantly better than control and Variant 2 which only had one purchase option. We also saw that 30% of users click on the ‘View more offers’ call to action on Variant 2.
  3. By just focusing on one product (All-access Digital) and surfacing monthly and yearly options, we were able to shift 40% of users to purchase the yearly term option.

What’s Next

  1. We’re conducting a follow-up test for Variant A removing the ‘View more offers’ call to action.
  2. We’re testing paywall designs with all product types on the paywall. Hypothesis: Users want to feel that they have explored and exhausted all possible purchase options.
Future designs to test — exposing all products in paywall
Future designs to test — exposing all products in the paywall

3. Pursue a new product feature: The ability for users to checkout within the paywall.

Future designs to test — checkout in paywall
Future designs to test — checkout in paywall