Analyze a 20 Percent Retail Revenue Decline
A retailer reports a 20 percent decline in revenue compared with a prior comparable period. Assume revenue is net of returns and cancellations, and the two periods are intended to be comparable in length and seasonality. If they are not comparable, explain how you would adjust.
Design an analysis to pinpoint the drivers behind the revenue drop.
Constraints & Assumptions
-
Start with measurement and comparability checks before diagnosing business causes.
-
Decompose revenue into traffic, conversion rate, and average order value.
-
Segment by product, channel, region, customer cohort, and time.
-
Quantify contribution to the decline rather than listing possible causes only.
Clarifying Questions to Ask
-
Which periods are being compared, and are holidays, promotions, or trading days aligned?
-
Is revenue gross or net of returns, discounts, taxes, shipping, cancellations, and foreign exchange?
-
Were there tracking, pricing, inventory, checkout, marketing, or site changes?
-
Is the decline broad-based or concentrated in specific segments?
Part 1 - Measurement and Revenue Bridge
Explain the first checks and the decomposition you would build.
What This Part Should Cover
-
Validate data pipelines, definitions, currency, refunds, attribution, and period alignment.
-
Decompose revenue as traffic times conversion rate times average order value, with contribution from each component.
-
Include order count, units, ASP, discounting, returns, and margin if available.
-
Build a bridge that accounts for the full 20 percent decline.
Part 2 - Segmentation and Diagnostics
Identify segments and root-cause areas to investigate.
What This Part Should Cover
-
Slice by product category, SKU, channel, region, device, customer cohort, acquisition source, and time.
-
Check inventory availability, pricing, promotions, marketing spend, competitor activity, site speed, checkout errors, payment failures, and fraud rules.
-
Compare new versus returning customers and paid versus organic traffic.
-
Use waterfall or mix-shift analysis to distinguish volume changes from mix changes.
Part 3 - Quantification and Causal Validation
Describe how you would validate which factors caused the decline.
What This Part Should Cover
-
Use experiments, diff-in-diff, matched controls, interrupted time series, or regression where appropriate.
-
Validate inventory or site-performance hypotheses with event timing and affected cohorts.
-
Estimate effect size and uncertainty for major candidate drivers.
-
Recommend immediate mitigations and follow-up instrumentation.
Follow-up Questions
-
What if traffic is flat but conversion rate falls sharply?
-
How would you analyze a revenue decline caused by mix shift toward lower-priced products?
-
What would you do if the largest decline is in a segment with missing tracking data?