Most stores do not have a traffic problem, they have a leak. Doubling ad spend on a funnel that converts at 0.7% just buys more expensive disappointment. This is the sequence we use to find and fix leaks, in the order that returns money fastest.
Measure revenue per visitor, not conversion rate
Conversion rate alone can rise while revenue falls - discounting does exactly that. Revenue per visitor combines conversion, average order value and traffic quality into one honest number.
Solution: instrument view, add-to-cart, checkout-start and purchase events cleanly, then read the drop-off between each step. The largest percentage drop is your first project, whatever your instinct says.
Product pages: answer the objection, then sell
Indian shoppers want sizing, ingredients, delivery timelines, return terms and proof that the brand is real. If any of that is missing, they open a competitor tab instead of asking.
Solution: put shipping and returns near the buy button, add real photography at usable zoom, show reviews with substance, and write specifications in plain language. This is usually worth more than any redesign.
Cart and checkout friction
Forced account creation, surprise shipping costs at the final step, and long address forms cause abandonment that no ad creative can rescue.
Solution: allow guest checkout, show delivery cost before the last step, autofill pin-code data, and keep the checkout to one screen on mobile. Abandoned-cart recovery over WhatsApp and email recovers a meaningful slice of the rest.
Mobile speed as a revenue lever
Most Indian ecommerce traffic is mid-range Android on mobile data. Every second before the first meaningful paint costs orders during peak sale periods.
Solution: budget page weight like money - lazy-load anything below the fold, defer non-critical scripts, compress imagery to WebP, and retest on a real device after every app install.
Test in a queue, not all at once
Changing hero, pricing, PDP layout and checkout in the same week means you learn nothing about which change worked.
Solution: keep a ranked test backlog, run one meaningful change at a time, and give each test enough orders to be believable before calling it.
- Track revenue per visitor as the primary scoreboard.
- Fix the biggest funnel drop-off first, not the prettiest problem.
- Delivery, returns and sizing information belong beside the buy button.
- Guest checkout and upfront shipping costs reduce abandonment immediately.
- Run one change at a time so results remain readable.