A month-over-month drop on Amazon is a seasonal dip if the same product fell by a similar share in the same months last year and the year-over-year figure for the current month is flat or up; it is a real decline if year-over-year is down too, and a trailing twelve-month line confirms the direction. Five tests settle it: year-over-year for the same period, the trailing twelve months, the category's known seasonality, an adjustment for days out of stock, and a split of the drop into traffic and conversion. Month-over-month alone cannot tell the two apart, and a seller who cuts ads or price on a seasonal dip does the damage himself.
Compare August to last August, not to July. Seasonality repeats on a twelve-month cycle, so the same month a year ago carries the same calendar. If this August is up on last August, the July-to-August drop is the calendar. If it is down, keep going. The period comparison glossary covers how to line up the periods when a month has a different number of weekends or a Prime Day moved.
Sum the last twelve months and compare to the twelve before. Seasonality washes out of a twelve-month sum because every season is in both windows. A product with a seasonal dip shows a flat or rising trailing line; a product in decline shows the line bending down, usually for three or four months before the operator notices.
Garden, pool, and grilling fall from August; school supplies spike in August and collapse in September; toys build from October. If the category has a known shape, a dip that matches the shape is presumed seasonal until tests one and two say otherwise. Amazon's own seller pages advise planning around seasonal storage rates and holiday demand, which is the same shape from the fee side.
A product that was out of stock for seven days in a 31-day month sold on 24 days. Gross up the month by 31 divided by 24 before comparing it to anything, or the stockout reads as a decline. Then check whether the comparison month had its own gap. Two months with different in-stock days are not comparable until both are adjusted.
Business Reports give sessions and unit session percentage by ASIN (sell.amazon.com describes both in its Business Reports guidance). Units equal sessions times conversion. If sessions fell and conversion held, the product lost visibility: rank, ad impressions, or the category's search volume. If sessions held and conversion fell, something on the page or in the market changed: price, reviews, a competitor, a suppressed image. Seasonal dips usually show up as falling sessions with steady conversion, because fewer people are looking. A conversion drop with steady traffic is rarely the calendar.
Illustrative numbers. A garden hose nozzle.
That is the number that starts the panic. Now the tests.
Year over year. Last July: $41,000. Last August: $33,800. Last year's July-to-August drop was 17.6 percent, within a tenth of a point of this year's. August this year against August last year: $39,600 divided by $33,800 is up 17.2 percent.
Trailing twelve months. $420,000 this year against $362,000 the year before, up 16.0 percent, with the line rising in ten of twelve months.
Category. Garden. August is the tail of the season.
Stockout. In stock 31 of 31 days both months. No adjustment.
Traffic and conversion. Sessions fell from 62,000 to 51,000; unit session percentage held at 7.4 percent both months. Fewer people looked; the same share bought.
Verdict: seasonal. The product is growing 17 percent a year and August is August. The correct action is to check that the ad budget steps down with demand so TACoS does not climb while sales fall, and to set the reorder for the spring build-up, not to discount.
Illustrative numbers. A phone case.
A similar-looking drop. Run the same tests.
Year over year. Last July: $21,500. Last August: $22,100. Last year August was up 2.8 percent on July. This year's August against last year's August: $18,700 divided by $22,100 is down 15.4 percent.
Trailing twelve months. $256,000 this year against $259,000 the year before, down 1.2 percent, and down in each of the last four months against the same months a year earlier.
Category. Phone cases have a device-launch cycle, not a summer cycle. Nothing in the calendar explains August.
Stockout. In stock 31 of 31 days both months. Not a supply gap.
Traffic and conversion. Sessions went from 9,800 to 9,900, flat. Unit session percentage fell from 8.2 percent to 6.9 percent. The same number of people looked; fewer bought.
Verdict: decline, and a specific kind. Traffic is intact, so rank and ads are not the cause. Conversion fell 1.3 points with the same traffic, which points at the page or the market: a price move by a competitor, a run of poor reviews, a listing change, or a new device that made the case look old. The action is on the detail page and the competitive set, this week, and the trailing line says it started four months ago.
Both examples are revenue. The decision needs contribution. A seasonal product whose August contribution held at 26 percent while revenue fell is healthy. A seasonal product whose August contribution fell to 12 percent because the ad budget stayed at July's level has an ad problem hiding under a seasonal dip. The SKU-level profit report puts the margin beside the revenue for the same period, which is the only way to see that the nozzle's TACoS went from 9 percent to 14 percent because spend did not step down with demand.
Run the tests in a comparison report that lines up the same period last year, the prior period, and the current one on a single basis, by SKU, with fees, ads, refunds, and COGS underneath the revenue line. Month-over-month on the Seller Central sales dashboard is where the question starts. It is never where the answer is.
The five tests are a twenty-minute exercise per SKU by hand, which means they get run on the products a seller already suspects and skipped on the ones that are quietly bending down. Crunch, the analytics AI inside ConnectBooks, takes "is this product declining or is it seasonal" and runs the comparison across the right periods (month over month, the same month last year, quarter over quarter) on reconciled sales, fee, ad, return, and COGS data, then names the products responsible and why. It works on the money side: sales, fees, ads, returns, inventory, and margins. The traffic-versus-conversion split comes from Business Reports and stays your job to lay beside it, and it cannot see the competitor's price cut or the review that went viral; it will show the sales drop by product and leave that cause to you. It also cannot rescue a comparison built on a stockout month you never flagged. Tell it the gap, and read the ranking with the calendar in front of you.
Running an e-commerce business comes with plenty of challenges, but ConnectBooks is here to make your life easier. With real-time insights, seamless integrations, and detailed tracking of your profitability and inventory, you can stay ahead of the game. Whether you’re selling on Amazon, Shopify, Walmart, TikTok or eBay, ConnectBooks helps you manage your finances with 100% accuracy and confidence, so you can focus on growing your business.
Ready to level up? Start making smarter, data-driven decisions every step of the way. Try ConnectBooks Free Today or Schedule a Demo
