I modelled a cross-device routing decision, compared three directions and carried the chosen approach into components, stakeholder alignment and an A/B-test plan.
Ordering decisionConceptual visualization, not production UI
RoleUX Designer
System3 ordering paths
Decision3 options compared
DeliveryComponents + test plan
01 / Problem
A routing decision disguised as a modal
Users could continue through different ordering tools depending on product and device. The modal sat at the intersection of customer intent, technical readiness and business goals. High exit behavior on desktop and frequent mobile misclicks showed friction, but a mobile experiment was not statistically significant and had documented test influences. The responsible response was deeper modelling, not a convenient success claim.
User need
Reach the right tool
The fastest route depended on product intent, device and the capabilities available in each editor.
Business need
Support channel goals
Conversion and app growth mattered, but neither justified making the customer choose without enough context.
Evidence quality
Read uncertainty honestly
Click behavior and experiment results were interpreted with their limitations instead of treated as automatic proof.
System boundary
Product + device routing
The core model had to decide which route to recommend while keeping alternatives discoverable.
02 / System model
Map the decision before drawing the interface
I built a product and device decision tree, benchmarked comparable patterns and made the routing logic explicit. That separated the underlying product choice from the visual treatment of the modal.
01Identify product intent
02Read device context
03Check editor readiness
04Recommend one route
05Keep alternatives accessible
This is an anonymized reconstruction of the documented decision model. Internal percentages, production screens and commercially sensitive routing details are intentionally withheld.
Explore the option comparisonThree directions, explicit criteria and stakeholder alignment03 / Options + tradeoffs
Three directions, not one polished answer
Option 1
Send users directly into one recommended ordering path. Lowest interruption, but the system must choose correctly.
Option 2
Expose two primary paths. More visible choice, but additional comparison effort at a sensitive point in the journey.
Option 3
Keep and improve the modal. Familiar structure, but continued interruption and risk of repeating the original friction.
Decision criteria
Customer effort, clarity, editor readiness, conversion, app growth, technical effort and access to alternative routes.
Alignment
Reviewed the recommendation with UX/UI, Development, Product Owner and Business Owner input before prioritization.
Product decision
Recommend the strongest path directly and preserve access to the wider ordering overview.
The chosen direction reduced unnecessary interruption without pretending that one route fit every product and device context.
04 / Delivery + learning
Make the decision testable
01Analyze behavior
02Compare options
03Align stakeholders
04Specify components
05Plan the A/B test
The mature outcome was a defensible, buildable and testable product decision.
Wireframes, mobile and desktop components, prioritization and an A/B-test plan carried the model into delivery. Because the available experiment evidence included non-significant results and test influences, this case does not claim a fabricated conversion impact.
