Espresso paywall redesign
How a three-iteration approach combining competitive analysis, a cross-functional content workshop, and two rounds of usability testing grew free trial starts by over 50%, increased the annual subscription mix from 10% to 25%, and raised CLTV per user by 12%.
MY ROLE: Lead UX/UI designer · Owned all design decisions, workshop facilitation, research direction
OUTCOME: +50% trial starts · 10→25% annual mix · +12% CLTV per average user
TIMELINE: A bit over two sprints
TEAM: Espresso app team, Marketing, Legal, Engineering, Content design
BUSINESS GOAL: Improve conversion
I owned UX and UI end-to-end across all three iterations, from initial audit and competitive analysis through to workshop facilitation, design variants, usability testing, and final handoff. Content design was a collaborator on copy; I set the direction and they refined it. Legal was a key dependency throughout, particularly around auto-renewal and trial disclosure requirements.
Context
The business knew Espresso app subscription starts were lower than expected. We were asked to identify where the journey was breaking down. I took ownership of auditing the end-to-end subscription journey, reviewing competitor paywalls, and analysing The Economist's own web paywall to find where users were losing confidence or motivation to convert.
The analysis pointed clearly to the paywall as the primary point of failure. I brought three specific problems back to the team, each with a hypothesis to test, and recommended the paywall redesign as the intervention most likely to move the conversion needle.
Problems & hypotheses
1. Incomplete information
Problem statement: Pricing clarity, free trial terms, legal requirements, and product benefits are all either absent or poorly surfaced, creating uncertainty at the most critical moment in the journey.
Hypothesis: Surfacing complete pricing, trial terms, and benefits on the paywall would reduce uncertainty and increase free trial starts.
2. No annual subscription option
Problem statement: The web platform offered an annual subscription with proven results. The app offered only monthly, removing a higher-value path and limiting long-term revenue per subscriber.
Hypothesis: Introducing an annual option would increase the Espresso cross-platform annual mix and raise CLTV per user, mirroring web platform results.
3. Subscription commitment pressure
Problem statement: The paywall led with a direct subscription ask rather than a free trial offer, raising the perceived cost of trying the product for users who were interested but uncertain.
Hypothesis: Leading with the free trial offer rather than a direct subscription ask would lower the perceived commitment barrier and convert more hesitant users.
*Why the web paywall was the right reference point: The Economist and Espresso share a brand and subscriber base. Maintaining coherence across both subscription surfaces was as important as the design patterns themselves — inconsistency between the two would erode trust at the point of purchase.
The approach
Three iterations with a clear purpose each
I structured the work into three sequential iterations rather than attempting to solve everything at once. Each iteration had a defined scope and a specific reason for being separate.
First iteration
Getting the annual option live, fast
The first iteration had one goal: introduce the monthly and annual subscription choices as clearly as possible, within a tight deadline, and with minimal engineering complexity.
I made a deliberate decision to keep iteration 1 lean, showing prices only, no savings labels, no benefits copy, knowing the next iteration would add that layer once content alignment had been reached. Trying to get everything right in one pass would have missed the deadline and delayed the annual option going live.
I used radio buttons as the selection pattern, based on the competitive analysis and the web paywall precedent. Early designs went through review sessions with the design team, stakeholders, and Engineering — the feedback loop caught layout issues before they became build problems.
Second iteration
Collaborative workshop: shaping essential content
While Engineering built iteration 1, the content designer and I planned and ran a workshop with stakeholders from Marketing and Legal. I ran this before designing anything for iteration 2 — the paywall sits at the intersection of commercial messaging, legal compliance, and user experience, and getting the content priorities wrong would have meant designing the wrong thing. The workshop aligned the group around a content hierarchy that would guide all subsequent design decisions:
Primary information: Pricing options and communicating the free trial process — what you get, for how long, and what happens next.
Secondary information: Benefits of subscribing, the value of the annual option, and comprehensive terms covering auto-renewal and cancellation.
The workshop also surfaced an important consistency question: how closely should the Espresso paywall align with The Economist web paywall? The group agreed on a principle, broadly consistent in structure and tone, but adapted for the app context rather than copied directly.
Two design variants for usability testing
Drawing from the workshop output, I designed two variants for user testing, each applying the agreed content hierarchy but with a different primary framing. I collaborated closely with the Economist app designer throughout this phase, who was working on a parallel paywall challenge. That cross-team exchange meant both projects benefited from shared findings rather than solving the same problem in isolation.
Variant 1 — subscription framing: Title emphasised subscription promotion. Benefits presented as a list of what you get with Espresso.
Variant 2 — free trial framing: Title led with the free trial offer. App features surfaced to help users understand the product before committing.
What usability testing found
We tested both variants against the existing paywall. The findings were clear:
Variant 2 outperformed on clarity and conversion intent. The "start your free trial" framing reduced the perceived risk of signing up, and presenting app features helped users who were unfamiliar with Espresso understand what they'd be getting.
The legal copy additions (which some had been nervous about including) were actively welcomed by users as reassuring rather than off-putting.
Most users said they'd start with a monthly subscription and only consider switching to annual once they'd decided they liked the product. Rather than treating this as a failure, I reframed it as a sequencing insight. The paywall needed to make the monthly option feel like a low-risk entry point, with the annual option clearly positioned as the better long-term value once trust was established.
Informed by user testing insights, we identified a winning design variant. After presenting the research findings, we sought input from stakeholders, leading to unanimous approval and final sign off. The culmination of this journey marked the launch of the transformed paywall in July 2023.
Third iteration
Hierarchy, polish, and setting up for A/B testing
The third iteration addressed the remaining UX and UI improvements that would sharpen the paywall's clarity and set up two planned A/B tests. I worked with Engineering to confirm technical feasibility for all changes before finalising the scope — everything was buildable.
UX and hierarchy improvements
Annual and Monthly labels added to pricing options
Annual savings shown as both a monthly equivalent and a percentage saving (validated by a colleague's independent research on a similar problem)
Sticky "Start free trial" CTA to keep the primary action always visible
Log in link repositioned for consistent visibility
T&Cs and Privacy links moved adjacent to the CTA as per Legal requirements
A/B test 1 — copy changes
Testing title and pricing option copy changes requested by senior stakeholders, to measure impact on free trial starts. Test strategy scoped with the research team.
A/B test 2 — feature placement
Testing whether showing pricing options before the app features list (vs. after) increases subscription decisions. Placement order hypothesised to affect whether users commit before or after seeing feature detail.
The results
+50% increase in free trial starts. This increase in trial starts validated the core hypothesis that the original paywall was suppressing conversion through lack of information, not lack of intent.
10→25% annual subscription mix. The annual subscription mix moving from 10% to 25% was the commercial headline: that shift directly improves long-term revenue per subscriber
+12% CLTV per average user
The two A/B tests were planned for January 2024 to continue optimising copy and feature placement.
What I'd do differently
The three-iteration structure worked well in practice, but iteration 1 was more constrained than I'd have liked. Shipping with prices only and no savings information meant some users in that window were making a subscription decision without the full picture. I accepted that trade-off to meet the deadline, and the plan was always to improve it quickly, but in hindsight I'd push harder to include at least the savings label in the first release. It was a small addition with disproportionate value to the user.
The usability testing finding on monthly-first preference was the most important insight in the project and it arrived late, in iteration 2. Running even a lightweight piece of attitudinal research at the start, before any design work, might have surfaced that pattern earlier and shaped how I framed the annual option from the beginning.
* What I'm most confident about: running the content workshop before designing iteration 2. It would have been faster to make assumptions and iterate but the paywall is a high-stakes surface where legal, commercial, and UX needs genuinely conflict. Getting those stakeholders in a room to agree on a content hierarchy before touching Figma saved multiple rounds of late-stage feedback and meant the designs that went to user testing were already grounded in cross-functional agreement.