Order conversions are dropping for Flipkart?

I’d first verify how conversion is measured, identify where users are dropping off, and connect that change to a specific cause. The key is to understand what prevents an interested shopper from completing a purchase.

1. Clarify the product and scope

Flipkart’s core e-commerce experience allows customers to discover products, compare options, and place orders through its website and mobile app.

For this case, I’d focus on that shopping experience, excluding other businesses and services. I’d initially include all product categories, then segment the data to see whether particular categories explain more of the decline.

2. Define the metric and quantify the change

I’d clarify whether conversion means orders per visitor, purchasing users per visitor, or sessions that result in a purchase. These measures aren’t interchangeable because one customer can place multiple orders.

For this investigation, I’d propose:

Purchase conversion rate = Unique visitors who place at least one confirmed order ÷ Total unique visitors × 100

I’d keep the measurement period, purchase window, and user-identification method consistent.

The decline from 30% to 20% is 10 percentage points, or a 33.3% relative decline.

For illustration, with 1,000 visitors, that means a reduction from 300 purchasing users to 200. Actual order volume would also depend on traffic and orders per purchaser.

Before diagnosing the experience, I’d verify:

• Whether the metric definition or attribution window changed.

• Whether tracking, dashboards, and data pipelines are working correctly.

• Whether order records support the reported decline.

• Whether seasonal sales, holidays, or comparison periods explain part of the change.

The case confirms that measurement is unchanged and the decline has developed gradually over six months without an established seasonal explanation.

3. Segment the decline and check the context

I’d compare conversion across mobile and web, Android and iOS, regions, new and returning users, product categories, and acquisition channels.

The case tells us the decline spans platforms, regions, and user groups. That points toward something shared across the shopping experience.

I’d still inspect traffic composition. More visitors from low-intent campaigns could reduce overall conversion even if the experience within each segment stayed unchanged.

Next, I’d check external influences:

• Competitor pricing, promotions, new offerings, or new entrants.

• Changes in regulation or operating conditions.

• Negative publicity or reduced trust.

• Strikes, delivery disruptions, weather events, or other unusual interruptions.

• Changes in household spending, income, work routines, or shopping habits.

In this scenario, competitors aren’t experiencing a similar decline, pricing remains competitive, and no major external disruption has been identified. I’d therefore prioritise internal causes while keeping those alternatives open.

4. Locate the problem in the shopping journey

I’d map the journey as:

Visit → Browse or search → View a product → Add to cart → Proceed to checkout → Confirm order

I’d review recent interface changes, experiments, releases, marketing messages, pricing presentation, and business-policy changes against this funnel.

I’d verify both volumes and rates, and account for customers who buy directly rather than follow the cart journey.

The important finding is that users still discover products and add them to their carts. Something at the cart stage appears to change their decision.

5. Investigate why users abandon their carts

I’d start with three hypotheses from the case:

I’d also check cart errors, delivery charges, stock availability, and delivery estimates to avoid overlooking another explanation.

The interviewer identifies a price mismatch: in this scenario, tax is added in the cart after users have seen a lower price earlier.

I’d frame this as a change in tax presentation within the product, rather than assume a newly introduced tax regime. The relevant product issue is that users discover a higher payable amount late in their decision.

For example, a shopper might choose an item displayed at ₹1,000, then see ₹1,180 in the cart. These are illustrative amounts. Even if the final price is competitive, the unexpected increase can make the purchase feel less affordable or undermine confidence in the displayed price.

6. Validate that the mismatch explains the decline

I’d compare cart abandonment before and after the pricing-display change, particularly for products with different gaps between the initial price and the cart total.

I’d look for:

• Higher abandonment when the price increase is larger.

• More exits immediately after the total becomes visible.

• More price-related complaints or feedback.

• A stronger decline among users exposed to the changed presentation.

• Recovery when consistent pricing is restored.

I’d also reconcile the timing. A single release does not automatically explain a gradual six-month decline. I’d check whether the change expanded across categories or users over time. If it was introduced everywhere at once, I’d investigate why the effect continued worsening.

My leading diagnosis would be that late disclosure of the full price creates surprise at the cart stage and reduces checkout progression.

7. Fix the immediate problem and prevent recurrence

In the short term, I’d make the price shown before adding an item to the cart consistent with the price users are expected to pay, including applicable tax.

Any charges that depend on delivery location or another selection should be explained early and updated as soon as that information is available. The cart should provide a clear breakdown, with discounts reflected consistently.

I wouldn’t begin by adding bigger discounts. That could reduce margins while leaving the underlying trust problem unresolved.

For the longer term, I’d:

• Use consistent pricing calculations across listings, product pages, and checkout.

• Test price consistency across categories, quantities, locations, and promotions.

• Monitor unexpected differences between displayed and payable prices.

• Evaluate pricing designs through completed purchases and customer understanding.

• Review competitor experiences to identify useful patterns for presenting totals.

I’d also inspect sale messaging in the cart. If it encourages customers to postpone an otherwise intended purchase, that deserves a separate test.

8. Measure whether the change helps

My primary outcome would remain purchase conversion using the agreed definition.

Supporting metrics would include:

• Cart-to-checkout and checkout-to-order conversion.

• Cart abandonment.

• Price-related complaints.

• Unexpected price-mismatch frequency.

• Time from adding an item to completing the purchase.

I’d monitor revenue and contribution margin per visitor, cancellations, returns, and payment success as guardrails.

Showing the full price earlier might reduce add-to-cart rates because some shoppers recognise sooner that an item exceeds their budget. I’d judge the change by completed purchases and customer understanding across the full journey.

My recommendation is to restore consistent, transparent pricing throughout the shopping experience. The evidence points to a cart-stage price surprise, and the strongest validation would be improved purchase conversion after that surprise is removed.