Key Takeaways
- A window product drives ad clicks and catalog traffic but generates most of its value through cross-sell purchases that standard attribution tools including Sellerboard cannot see, because Amazon only reports the advertised ASIN in default attribution.
- One seller nearly paused a campaign showing 62 percent ACoS that was generating $17,000 in annual profit through accessory cross-sells invisible to their dashboard.
- The free Purchased Product Report in Seller Central reveals both the advertised ASIN and the actual purchased ASIN, exposing cross-sell revenue that transforms apparent losers into campaigns running at 7x or higher true ROAS.
- Categories with the highest cross-sell attachment rates include Beauty at 40 to 60 percent, Kitchen at 35 to 50 percent, and Electronics at 25 to 40 percent.
- True ROAS for a window product equals direct revenue plus cross-sell revenue divided by ad spend: a campaign that showed negative $72 monthly in Sellerboard produced positive $1,362 monthly in true profit on the same data.
- A 5-stage fix covering identification, measurement, protection, scaling, and re-measurement over 3 months restructures how campaign profitability is evaluated and prevents high-value window campaigns from being cut on misleading data.
General Summary
Amazon sellers using standard profit tools like Sellerboard face a critical attribution blind spot: window products that drive ad clicks but generate most of their catalog value through cross-sell purchases that default attribution cannot track. One seller nearly paused a campaign showing 62 percent ACoS that was generating $17,000 in annual profit through accessory sales invisible to their dashboard. The free Purchased Product Report inside Seller Central reveals this hidden revenue by displaying both the advertised ASIN and the actual purchased ASIN, exposing cross-sell patterns that transform apparent losers into high-ROAS campaigns. Categories with the highest attachment rates include Beauty at 40 to 60 percent, Kitchen at 35 to 50 percent, and Electronics at 25 to 40 percent. A 5-stage framework covering identification, measurement, protection, amplification, and re-measurement over 3 months allows brands to capture true campaign profitability and stop cutting their most effective traffic sources based on incomplete data. Amazon’s default attribution was built to track one click and one sale. That model does not reflect how customers actually shop.
Extractive Summary
A seller paused a campaign showing 62 percent ACoS in Sellerboard and their organic sales dropped 60 percent overnight. The Purchased Product Report, free and built into Seller Central, is the tool that reveals the attribution gap standard dashboards cannot see. A chef’s knife campaign showing a $72 monthly loss in Sellerboard produced $1,362 in monthly profit when cross-sell revenue was counted. Not every product is a window product: Beauty categories show 40 to 60 percent attachment rates while books and single-use consumables show minimal cross-sell. A 5-stage framework over 3 months transforms campaign profitability evaluation through identification, measurement, protection, amplification, and re-measurement. Amazon’s default attribution was built to track one click and one sale, but customers browse, compare, and buy multiple products from a single ad interaction.
Abstractive Summary
Amazon advertising attribution has been misaligned with actual customer shopping behaviour since the platform’s early days, creating a systemic blind spot that compounds in cost as catalogs grow more interconnected. The gap between standard attribution models and true purchase behaviour is widest in categories where accessory and complementary product ecosystems drive significant revenue. What profit tools report as underperforming campaigns often represent the highest-leverage traffic sources in a catalog, functioning as customer acquisition engines rather than direct conversion vehicles. This attribution failure reinforces itself over time: sellers systematically defund their most effective campaigns while over-investing in products that appear profitable only because their revenue was misattributed. Sellers who identify window products and protect them from standard optimisation decisions gain a structural advantage that is largely invisible to competitors still relying on dashboard ACoS alone.
Why Did Pausing a 62 Percent ACoS Campaign Kill 60 Percent of Organic Sales?
A seller paused a campaign showing 62 percent ACoS in Sellerboard. Their organic sales dropped 60 percent overnight. That campaign was generating $17,000 in annual profit. Sellerboard could not see it. The seller almost eliminated their most profitable traffic source based on a number that was completely wrong.
This pattern appears in every catalog audit. Products that look unprofitable in Sellerboard, Perpetua, and even Amazon’s Campaign Manager generate 2x, 3x, sometimes 7x more revenue than the dashboards show. The reason is window product attribution, and almost nobody checks for it.
What Is a Window Product and Why Can’t Profit Tools Track It?
A window product gets the click and the attention but not always the purchase. Think of a retail store window: the featured item draws people inside, but customers leave with the displayed item, shoes, a bag, and accessories. The window did its job. Measuring only window sales would make it look like it was failing.
Sellerboard calculates profit by a simple chain: customer clicks ad, customer buys that product, Sellerboard subtracts fees and ad cost. Clean logic. Completely wrong for roughly 30 percent of most catalogs. Sellerboard cannot see what else the customer bought after clicking.
How Does Cross-Sell Attribution Break Your Dashboard Numbers?
A seller running Sponsored Products ads for a chef’s knife sees this in their data: a customer clicks the knife ad, then buys the knife, a sharpening stone, and a knife sheath. Three products. One click. Sellerboard records only the knife sale. The sharpening stone and sheath appear as organic sales with zero ad cost allocated, as if they sold themselves.
They did not sell themselves. They sold because the knife ad brought that customer into the catalog. The knife campaign shows 62 percent ACoS. The sharpener and sheath show 0 percent ACoS. In reality, the knife campaign drove all 3 sales. True ACoS across the 3 products is closer to 12 percent.
What Happens When You Pause the Window Product?
The data appears to recommend a clear action: the knife is losing money, the sharpener is pure profit, pause the knife ads and let the organic sharpener sales continue. Those sharpener sales are not organic. They are 100 percent dependent on knife ad traffic.
One seller ran this exact experiment. They paused the knife ads. Sharpener sales dropped 60 percent in two weeks. Eighty percent of sharpener revenue had been coming from the knife click halo. A seller in the Amazon forums described it directly: Sellerboard showed 62 percent ACoS, they were ready to kill the campaign, ran the Purchase Product Report first, and found actual ROAS was 5.2x because of cross-sell. Trusting Sellerboard would have cut their most profitable campaign.
How Do You Access the Purchased Product Report in Seller Central?
Amazon provides the data to close this attribution gap at no cost. The Purchased Product Report is free, built into Seller Central, and used by almost nobody. Go to Seller Central, open the Reports menu, select Advertising Reports, click Create Report, choose Purchased Product as the report type, set the date range to 30 days minimum or 90 days for seasonal products, run the report, and download the CSV.
What Two Columns Reveal Your Hidden Revenue?
The report contains 2 critical columns: Advertised ASIN and Purchased ASIN. Advertised ASIN is the product the ad showed. Purchased ASIN is the product the customer actually bought. When those 2 columns contain different values, that is the window product effect in the data.
For the knife example: row 1 shows Advertised ASIN chef’s knife, Purchased ASIN chef’s knife, a direct sale that Sellerboard captures. Row 2 shows Advertised ASIN chef’s knife, Purchased ASIN sharpening stone, a cross-sell that Sellerboard misses. Row 3 shows Advertised ASIN chef’s knife, Purchased ASIN knife sheath, another cross-sell that Sellerboard misses. Four sales, one ad click, one sale counted.
How Do You Identify a Window Product in Your Catalog?
Divide total cross-sell revenue by total revenue from that campaign. If the result exceeds 50 percent, the advertised product is a window. It is a traffic driver, not the profit centre. This is data that Sellerboard literally cannot access, not because Sellerboard is a poor tool, but because it pulls from Amazon’s standard attribution, which only counts the advertised ASIN. The Purchased Product Report shows everything that happened after the click.
How Did a Losing Campaign Generate $16,344 in Annual Profit?
A chef’s knife campaign showed a $72 monthly loss in Sellerboard. The Purchase Product Report showed $1,362 in monthly profit. Same campaign. Same underlying data. Completely different conclusion.
What Did Standard Attribution Miss in the Knife Campaign?
The campaign numbers: $500 in monthly ad spend, 1,000 clicks, 120 total conversions. Sellerboard saw 35 knife sales at $60 each, producing $2,100 in revenue. Revenue of $2,100 on $500 spend equals 4.2x ROAS. After Amazon fees and cost of goods, the campaign showed negative $72 per month. Standard attribution said to cut it.
The Purchase Product Report showed the same 120 conversions broken down differently: 35 knife sales at $2,100, 45 sharpening stone sales at $1,125, 28 knife sheath sales at $420, and 12 cutting board sales at $285. Total revenue from that $500 in ad spend: $3,930.
What Is the True ROAS Formula for Window Products?
True ROAS equals direct revenue plus cross-sell revenue, divided by ad spend. For the knife campaign: ($2,100 + $1,830) divided by $500 equals 7.86x. Not 4.2x.
Standard attribution produced: $2,100 revenue, $500 ad spend, $1,672 in fees and cost of goods, negative $72 net profit. True attribution produced: $3,930 revenue, $500 ad spend, $2,068 in fees and cost of goods, positive $1,362 net profit. Same campaign, same month, $1,434 difference in measured profitability.
What Is the Annual Impact of Wrong Attribution Decisions?
Trusting Sellerboard on this campaign means pausing it, saving an apparent $72 monthly loss, which is $864 annually recovered. Trusting the Purchase Product Report means keeping it running, generating $1,362 monthly, which is $16,344 annually in profit. The swing between those 2 decisions: $17,208 per year on one campaign.
A second example shows the same pattern in electronics. An earbuds campaign showed 50 percent ACoS, appearing break-even at best. The Purchase Product Report revealed the earbuds were driving webcam sales: customers clicked the earbuds ad and bought an $80 webcam instead. Fifty earbud sales at $2,000 plus 150 webcam cross-sales at $12,000 produced $14,000 in total revenue on $1,000 in ad spend. True ROAS: 14x. True ACoS: 7.1 percent. The earbuds were not the product. They were the window.
Which Product Categories Have the Highest Cross-Sell Rates?
Not every product functions as a window. Some products sell themselves as standalone destinations. Others exist primarily to bring traffic into the catalog. Category data reveals which is which, and the difference determines whether standard attribution or Purchase Product Report data should govern campaign decisions.
What Are the Highest and Lowest Cross-Sell Categories?
Beauty categories including shampoo and skincare show 40 to 60 percent attachment rates. A customer clicking a shampoo ad frequently buys conditioner, treatment, and styling product in the same session. Shampoo is almost always functioning as a window in a multi-SKU beauty catalog.
Kitchen products including knives and cookware show 35 to 50 percent attachment. A knife drives sharpener, block, and cutting board sales. A pot drives lid, utensil, and accessory purchases. Electronics including earbuds and cables show 25 to 40 percent attachment. Apparel including shoes and basics shows 20 to 35 percent attachment.
The lowest cross-sell categories are books, single-use consumables, and highly specialised tools. These are destination products: customers click, buy that specific item, and leave. Standard attribution is reliable for destination products. It is not reliable for categories with natural accessory ecosystems.
How Do You Know If Your Product Is a Window or a Destination?
A product with natural accessories or companion items in the same catalog is likely a window. A standalone product with no related catalog items is likely a destination. Windows require the Purchase Product Report to evaluate accurately. Destinations can trust standard attribution.
What Is the 5-Stage Fix for Window Product Attribution?
This 5-stage framework applies to every brand with a multi-SKU catalog. Run through all 5 stages over 3 months to restructure how campaign profitability is evaluated and protected.
How Do You Identify and Measure Window Products?
Stage 1, Identify (Week 1): Pull the Purchase Product Report for the last 90 days. Sort by Advertised ASIN. Flag any ASIN where cross-sell revenue exceeds 30 percent of total revenue from that campaign. Those are the windows.
Stage 2, Measure (Week 2): Calculate true ROAS for each window product using the formula: direct revenue plus cross-sell revenue divided by ad spend. Any campaign where true ROAS exceeds the target threshold stays on, regardless of what the standard dashboard shows.
How Do You Protect and Scale Window Product Campaigns?
Stage 3, Protect (Week 3): Tag window campaigns internally and document the cross-sell ratios. No decisions are made on these campaigns without first checking the Purchase Product Report. No optimisation based on standard ACoS alone.
Stage 4, Amplify (Month 2): If a window product is driving 7x true ROAS, scale it. Increase budget 20 to 30 percent and test additional keywords. The cross-sell effect typically scales proportionally with traffic volume. More window traffic means more catalog sales.
Stage 5, Measure Again (Month 3): Re-pull the Purchase Product Report. Verify the cross-sell ratios have held through the scaling period. Some window products are seasonal. Others are evergreen. The data distinguishes them.
What Are the Four Automation Options for Cross-Sell Tracking?
4 options exist for operationalising cross-sell attribution at different cost and complexity levels. Manual Allocation requires 2 to 3 hours monthly and costs nothing: calculate attribution percentages, allocate ad spend proportionally, and recalculate profit and loss manually. Campaign Bundle Analysis requires 1 hour monthly and costs nothing: stop evaluating individual ASINs and evaluate campaign ecosystems instead, dividing total revenue from all products by total ad spend. Most brands start here.
Amazon Marketing Cloud costs $900 to $5,000 monthly and runs SQL queries identifying exact cross-sell patterns with multi-touch attribution. Highest accuracy, highest cost. Custom BI dashboards using Tableau or Looker cost $1,500 to $3,000 monthly and automate Purchase Product Report imports into custom profitability views. The right option depends on catalog size and the revenue at stake in window campaign decisions.
Why Does Amazon Attribution Fail to Show Your True Profitability?
Amazon’s default attribution was built to track one click and one sale. That model does not reflect how customers shop. They browse, compare, add multiple items to cart, and buy 3 things because an ad showed them one. Sellerboard and similar tools can only see what Amazon’s standard attribution reports. Standard attribution does not report cross-sell.
The campaigns that appear to be failing may be funding the entire catalog. The Purchase Product Report is the tool that makes that visible. Stop optimising for the click. Optimise for the customer.
