A Bash-shaped commerce model built to show how acquisition, checkout, fulfillment, and returns interact. The goal is not a prettier dashboard. The goal is a decision view that product, growth, and operations can act on together.
Of every 100 sessions, only 6.7 convert to purchase. Product-stage exits alone absorb 59.0% of all non-converting sessions.
App, desktop, and mobile web through the same five-step journey.
Each stage shows its share of total dropoff and the biggest reason inside that stage.
Top reason: Landing bounce (38.6% of this stage).
Top reason: Size uncertainty (30.9% of this stage).
Top reason: Delivery cost shock (34.8% of this stage).
Top reason: Payment failure (34.5% of this stage).
The same catalogue performs very differently across surfaces. That makes the mobile web gap a conversion problem, not a traffic quality problem.
Three surfaces, three very different outcomes.
Completion rate among sessions that already reached checkout.
Late orders make up 12.1% of the model, but they lift support contact to 31.1% and return rate to 21.6%.
Support rate, return rate, and delivery days split by outcome.
One delivery slip creates pressure far beyond logistics.
Footwear carries the highest modeled return rate at 22.9%. The strongest signal inside the return mix is size issue.
Footwear leads the pressure profile, with Women Apparel close behind.
Return volumes point directly to content and expectation gaps.
Ordered by practical commercial impact. Every move below is grounded in a visible signal from the modeled journey.
Halfway to app performance would add about 233 purchases from existing sessions.
Split deliveries push support contact to 16.8% versus 7.7% on single-parcel orders.
Only 12.1% of orders are late, but they trigger a 31.1% support-contact rate.
22.9% return rate with size issue leading the return mix.
This case study uses synthetic but behaviorally consistent data designed to reflect an omnichannel retail workflow. It is a decision model, not a claim of access to internal TFG data.