Checkout Funnel Optimization
How session recording analysis revealed two hidden friction points killing checkout conversions — and fixing them contributed to ₹1.2 crore in monthly revenue improvement.
Company
Sierra Living Concepts
Timeline
Nov 2025 – Jan 2026
My Role
Product Manager
Teams
UI/UX · Category · IT
The journey, before → after
Added to cart
Distracted by similar items
Popup loses entered data
Redesigned — order confirmed
~₹1.2Cr
Monthly revenue improvement
20→14%
ATC diversions reduced
84→72%
Checkout abandonment
The Background
In my early weeks as Product Manager, I made it a habit to review Microsoft Clarity session recordings and BI dashboard funnel reports every morning. Sierra's average order value is $3,500–$4,000 — meaning every customer who drops off the checkout flow represents thousands of dollars in lost revenue.
The business was generating reasonable traffic to product pages, and add-to-cart rates were acceptable. But somewhere between "add to cart" and "order confirmed," we were losing people. I needed to find out exactly where and why.
Discovery — What the Data Showed
Combining Clarity session recordings with BI dashboard funnel analysis, I mapped the full ATC-to-checkout journey and identified two distinct friction points:
Funnel drop-off map
Before vs After
Before
After
Checkout abandonment rate
11.32 point improvement
Requirements Definition
| Requirement | Description | Priority |
|---|---|---|
| Remove ATC Similar Products | Remove distracting similar product recommendations from ATC page to eliminate 20% diversion | High |
| Dedicated Checkout Step | Replace popup with dedicated next-step page — separate URL, no overlay | High |
| Form Session Persistence | Entered form data must persist if user navigates away or leaves idle — no data loss | High |
| Retain Complementary Cross-sell | Keep relevant cross-sell (e.g. dining chairs on dining table ATC) — relevant, not distracting | Medium |
| Mobile Checkout Optimization | Dedicated checkout step fully optimized for mobile — sticky CTA, large inputs, minimal scroll | Medium |
Sprint Planning & Execution
Sprint Goal: Quantify the full funnel impact of both friction points and align UI/UX, category, and IT teams on the solution approach.
Sprint Goal: Remove similar product distractions from ATC page and ship the UI/UX design for the new dedicated checkout step.
Sprint Goal: Build the dedicated checkout step page, implement form data persistence, QA across device types, and deploy to staging.
The Key Product Decision
One deliberate decision I made that isn't obvious: I kept complementary cross-sell on the ATC page while removing similar product recommendations.
"Dining chairs on a dining table ATC page adds value to the existing decision. Similar dining tables create doubt about it. The distinction between helpful and distracting cross-sell is context, not category."
Results
20→14%
ATC page diversions — users leaving for similar products
↓ 6 percentage point reduction
84→72%
Checkout abandonment rate after info form redesign
↓ 11.32 percentage point improvement
~₹1.2Cr
Monthly revenue improvement (700 orders × $3,500 AOV × 6% recovery)
↑ From baseline
0
Customer-reported data-loss incidents post-deployment
Bug fully resolved
What I Learned
Session recordings reveal what metrics can't.
The BI dashboard showed abandonment numbers. Clarity showed exactly what users were doing — clicking similar products, tabbing away, coming back to an empty form. Qualitative and quantitative together tell the complete story.
Not all cross-sell is equal.
Removing similar products improved conversion, but we kept complementary cross-sell. Context-appropriate recommendations help; distraction disguised as recommendation hurts.
Bugs in the purchase flow are the most expensive bugs.
The popup form data-loss issue had probably been losing orders for months. Customer feedback was the signal — building a habit of reading support tickets surfaces product issues faster than any dashboard.