Plan Creation Flow Revamped
Client
Subsbase
Year
2022
This project redesigned Subsbase's plan creation flow, the core authoring surface of a subscription-billing platform serving merchants across MENA. The original flow had shipped as a single long-form page, and internal concern about drop-off, editing friction, and adoption had been building inside the product team well before the redesign brief was written.
The redesign rebuilt the flow from task architecture up: replacing a monolithic collapsible form with a four-step guided wizard, introducing a dedicated plan profile for post-creation management, and restructuring the field grouping so merchants could create, review, and edit plans without fighting the interface at any stage of the lifecycle.
Scope of Work
Problem Statement
Subsbase's entire value proposition rested on making subscription management effortless, and yet the single action that activated that value, creating a plan, had quietly become the hardest thing to do on the platform. Merchants described the flow as overwhelming, abandonment mid-creation was common, and edits after launch were reported as worse than the creation itself.
Underneath the frustration sat three structural problems the original form could not solve without a rebuild:
Flat architecture without progression. Every configuration section, sat stacked on one page behind collapsible rows. Users had to expand each section manually with no sense of how far they were from finishing.
Endless scroll with no orientation. Once sections were expanded, the page became an unbroken vertical column, and users lost track of which fields belonged to which section. Context was continuously rebuilt by scrolling.
Editing locked and hostile. After a plan was saved, inputs locked by default. Even small corrections forced users to re-expand sections, re-enter edit mode, and hunt for the field they needed, recreating the full creation burden for a minor change.
Goal
Success meant repositioning plan creation from a task merchants postponed into one they completed confidently on first attempt. Three goals anchored the project:
Simplify the flow into clear, manageable steps so users could create a plan without needing to understand the full configuration model upfront.
Reduce time-to-first-plan by surfacing only essential fields at creation, deferring advanced configuration to a dedicated post-creation surface.
Give every existing plan a clean, organized profile page so editing and managing live plans felt as intentional as creating them.
Research & Analytics
User interviews were run with existing clients who had previously created plans, and sessions included both fluent users and users who had abandoned the flow. Observation sessions captured where users paused, where they scrolled back, and where they gave up. Support tickets related to plan creation and editing were tagged and counted so the pattern would show up as frequency, not anecdote.
In parallel, competitor flows across Stripe Billing, Chargebee, and Recurly were audited to map the conventions merchants were increasingly expecting. Three patterns held across all of them: stepped creation with progressive disclosure, minimum-viable required fields upfront, and grouped post-creation editing under a tabbed or sectioned profile.
Design Thinking Process
Brainstorming workshops with product, design, and engineering prioritized features and set requirements. Constraints around legacy billing logic were surfaced early.
Wireframes explored step sequencing, field grouping, and navigation patterns for post-creation editing. Multiple iterations were shared with stakeholders to pressure-test flow against business goals and technical feasibility.
Final concept rested on two pillars:
A four-step wizard with only required fields upfront.
A plan profile page with top navigation bar and grouped sections.
Design Concept
The redesign was guided by the research findings, not by a style preference. Every design decision traced back to a specific complaint theme or observed drop-off point, which kept the critique loops focused on whether the flow was doing its job rather than whether the direction felt right.
Four-step plan creation wizard
Creation was broken into four linear steps, each focused on one aspect of the plan
Post-Launch Testing & Success Measurement
New design was tested with users who had previously struggled with old flow. Usability sessions measured ease of use, time to create a plan, and overall satisfaction. Feedback drove refinements to field grouping and profile navigation.
Post-launch, A/B testing was run against legacy flow. Follow-up interviews and surveys captured qualitative signal. Task completion rates were monitored to confirm users could create plans unassisted.
Results spoke clearly:
Efficiency: average plan creation time dropped by 60%. Task completion rose from 70% to 95%.
Satisfaction: 86% positive score in post-launch surveys.
Adoption: 85% of users moved to new flow immediately. Support tickets related to plan creation fell sharply.
Business impact: plans created per user increased 30% within three months. Retention improved as subscription management felt lighter.
Challenges & Conclusion
Not every trade-off was clean. Two areas asked harder questions than the pre-launch brief anticipated.
Legacy billing constraints could not be fully abstracted
Some field dependencies were tied to billing logic that predated the redesign, and a handful of fields remained non-editable on plans with active subscribers regardless of how the UI was restructured. Communicating these constraints clearly, without reintroducing the ambiguity of the original locked inputs, took several iterations of inline messaging and microcopy before the explanation felt honest rather than apologetic.
Splitting the task changed how users formed a mental model
The four-step wizard was unambiguously better for first-time creation, but a small group of power users, merchants who had internalized the old single-page form, initially reported that splitting the task felt slower even when measured completion time was faster. The lesson carried forward: the cognitive model a user builds around a flow is itself part of the design, and migrating long-term users requires patience even when the new flow is quantifiably better.
The Subsbase plan creation redesign validated the core bet: that a task drowning in its own configuration can be rescued by separating creation from management, surfacing only what is needed at the moment it is needed, and treating post-creation editing as a first-class surface rather than an afterthought. Every top-line metric moved positively, time-to-create fell 60%, completion rose to 95%, plans-per-user climbed 30%, and the flow that had quietly undermined the platform's value proposition became the clearest expression of it.
