Analysis of four monthly Business Reports covering April to July 2026, carried out to establish the cause of a sharp decline in sales before any corrective work began.
Determine whether the decline was driven by traffic, conversion or pricing, since each requires a different response and treating one as another wastes a quarter.
| Month | Sessions | Units | Sales | Conversion |
|---|---|---|---|---|
| April | 1,602 | 186 | $5,386 | 11.6% |
| May | 865 | 98 | $2,934 | 11.3% |
| June | 649 | 76 | $2,275 | 11.7% |
| July | 690 | 70 | $2,069 | 10.1% |
| April to July | -56.9% | -62.4% | -61.6% | -1.5 pts |
The decline is a loss of traffic rather than a failure of conversion. Sessions fell 56.9% while conversion held between 11.3% and 11.7% across three of the four months. A listing converting at 11% is performing above category norms, so the product converts and fewer people are reaching it.
The July conversion figure of 10.1% is a secondary effect worth monitoring rather than the cause. Had conversion fallen while sessions held, the response would have been listing content, pricing or reviews.
Concentration compounds the effect. The lead SKU accounts for roughly two thirds of sessions and a similar share of revenue, so a fall in its visibility moves the whole account.
The finding directed the remediation toward auction eligibility rather than listing quality. Upload 2 addressed exactly this: auto campaigns running $0.03 to $0.05 base bids were ineligible for most auctions, which is a loss of traffic with an advertising cause. A conversion-led reading would have sent the same eleven days into rewriting listings and images.