Amazon Sponsored Display looks like the calm, sensible part of Amazon Ads. You are not fighting for the top search slot. You are not rebuilding keyword match types. You are simply reminding shoppers who already showed interest, defending product pages and maybe reaching similar audiences off Amazon. Lovely. Also just dangerous enough to deserve adult supervision.
The named mistake I see in marketplace ad accounts is treating Sponsored Display as “cheap extra ROAS”. A team launches views remarketing, adds competitor ASIN targeting, sees a 7.8 ROAS, and quietly moves more budget into the format. Nobody asks whether the advertised SKU can afford the extra clicks. Nobody checks whether the sales were incremental. Nobody separates CPC campaigns from vCPM campaigns with view-through credit. And nobody notices that the product went out of stock three days later, so the campaign spent its last euros warming up shoppers for a weaker offer.
My stance: Amazon Sponsored Display should not be scaled as a channel. It should be scaled as a set of profit-permission plays. Views remarketing, self-targeting, competitor conquesting, purchase remarketing and vCPM reach each deserve different budgets, margin thresholds and evidence rules.
For brands spending from roughly €5K per month across Amazon, bol and MediaMarkt, this matters because Sponsored Display is usually the first “full-funnel” format that enters the weekly meeting. If you govern it only with ACOS, it can make the dashboard prettier while making the P&L worse. This guide shows how to run Sponsored Display with enough structure to protect contribution margin.
What the research gets right — and what it tends to miss
The competitive advice is useful, but incomplete. BidX explains the format clearly: Sponsored Display can appear on and off Amazon, can use views remarketing for shoppers who visited product pages in the last 30 days without buying, and can use product targeting for ASINs or categories. Their 2026 PPC guide also positions Sponsored Display as a way to retarget shoppers, cross-sell and defend detail pages when layered with Sponsored Products.
Amazon’s own documentation adds an important operational point that many sellers forget: display audiences are built for retail signals. Amazon highlights purchase remarketing, customizable lookback windows, product and category remarketing, bid optimization, and retail-aware protections such as out-of-stock, delivery-promise and Featured Offer awareness. It also notes that shoppers often take several days between starting research and purchasing, which is exactly the gap Sponsored Display tries to cover.
Jungle Scout gives a practical seller-friendly explanation: Sponsored Display does not target keywords in the same way as Sponsored Products. It targets audiences, product pages and shopping behaviour, and brand-registered sellers can use it for product targeting and views remarketing. Adference focuses on retargeting and the three targeting routes: audience interests, product targeting by product, and product targeting by category. Zane Marketing makes the sharpest point I found: Sponsored Display is “three products wearing one name”, and vCPM can inflate reported ROAS through view-through attribution.
Seller discussions on Reddit show the real hesitation behind the glossy guides. Sellers worry about limited placement control, off-Amazon spend they cannot fully inspect, and whether Sponsored Display deserves budget after Sponsored Products and Sponsored Brands. One common seller split is roughly 50–70% Sponsored Products, 20–30% Sponsored Brands and “whatever is left” for Sponsored Display. That is not a strategy, but it is an honest signal: operators do not fully trust the format until it proves itself.
The missing angle is profit permission. Most guides explain where ads show, how targeting works and which audiences to build. Fewer explain when a SKU is commercially allowed to receive retargeting budget, how to separate defensive spend from incremental spend, and when a high ROAS should still be cut.
The Sponsored Display map: five plays, five different rules
Before you touch bids, split Sponsored Display into five plays. A single campaign naming convention will not save you if these jobs sit in one budget bucket.
1. Self-targeting for detail-page defence
This is the cleanest play. You target your own ASINs so your other products, bundles or variants appear on your product detail pages. The job is not discovery. The job is to keep shoppers inside your catalogue instead of letting a competitor steal the carousel.
Use this for profitable hero products, multipacks and accessories. The budget rule can be modest: 5–8% of Amazon ad spend if the SKUs have stock cover and positive contribution margin after ads. Judge it by assisted basket value, advertised SKU margin and whether competitor clicks on your own PDPs drop over time.
2. Views remarketing for considered purchases
Views remarketing reaches shoppers who viewed a product detail page but did not buy. This is usually the first Sponsored Display campaign worth testing for products with a longer comparison cycle: electronics accessories, home appliances, premium toys, beauty devices, baby gear and higher-ticket hobby products.
Start with a 14- or 30-day lookback window. Seven days can be too narrow for categories where shoppers compare reviews. Ninety days can be too stale unless the product is expensive and research-heavy. The rule: if the SKU cannot carry a retargeting click at your expected conversion rate, it does not get scaled just because the audience is warm.
3. Competitor ASIN targeting for conquesting
Competitor targeting is tempting because it feels wonderfully direct: show your product on a rival’s detail page and capture demand at the decision point. The trade-off is brutal. You are often buying clicks from shoppers who already have a competing product in mind.
Only run this when your offer has a visible reason to win: better price, better review score, faster delivery, stronger bundle, or a spec the competitor lacks. If your product is €3 more expensive, has fewer reviews and ships slower, Sponsored Display is not conquesting. It is sponsoring your own rejection.
4. Purchase remarketing for replenishment and cross-sell
Purchase remarketing is excellent for consumables, refills, accessories and upgrades. It is weak for products people buy once every few years. A coffee filter brand can use a 45-day window to sell refills. A laptop stand brand should probably use it to sell a cable organizer or webcam light, not another identical stand next week.
5. vCPM reach and audience expansion
This is the grown-up testing area, not the default. vCPM can be useful for awareness, launches and upper-funnel support, but it changes what the report means. View-through attribution can make ROAS look healthier than click-only performance. Treat vCPM as a test with a holdout or on/off read, not as proof that profit appeared by magic.
The profit-permission checklist before launch
At FiveX we prefer boring permission rules over heroic weekly explanations. Before a Sponsored Display campaign goes live, each advertised SKU should pass five checks.
- Contribution margin: calculate margin after marketplace fees, fulfilment, payment costs, expected returns and current promo discounts.
- Break-even ACOS: set a campaign-specific ceiling below SKU margin unless the campaign has an explicit launch or defence role.
- Stock cover: avoid scaling retargeting if the SKU has less than 21–30 days of stock cover, unless replenishment is confirmed.
- Offer quality: check price position, reviews, delivery promise, Buy Box or Featured Offer status and product-page conversion readiness.
- Incrementality evidence: define whether the campaign must prove click-attributed sales, new-to-brand share, repeat purchase, PDP defence or on/off lift.
This is where FiveX naturally earns its seat in the workflow. The profitability dashboard brings SKU margin, ad spend, fees and returns into one view. Inventory insights show whether the product can actually absorb extra demand. Advertising automation can then apply rules that reduce bids, pause SKUs or move budget before the weekly meeting discovers the issue manually.
Named example 1: the €42 kettle that should not retarget forever
Imagine “NordKettle 1.7L”, an Amazon.nl kitchen product selling for €42. The landed product cost is €17.50. Amazon referral and fulfilment costs are €8.90. Expected returns and service costs are €1.60. That leaves €14.00 contribution before ads, or a 33.3% pre-ad margin.
The team launches Sponsored Display views remarketing with a 30-day window. Spend is €600 in two weeks, attributed sales are €2,850, and reported ACOS is 21.1%. On the surface, nice. The problem: Sponsored Products already captures most branded and category demand, and total Amazon sales for the SKU only rose from €9,800 to €10,450. TACOS moved from 11.8% to 15.2%.
The decision is not “ROAS is above target, scale it.” The decision is: cap views remarketing at €30 per day, test a 14-day window, and only scale if total SKU contribution grows by at least €350 in the next two-week period. In FiveX, this would be labelled Warm audience — capped until incremental lift, not “winning campaign”. Slightly less glamorous. Much more profitable.
Named example 2: the MediaMarkt accessory lesson for Amazon conquesting
A consumer electronics brand sells a “VoltCase USB-C Hub” on Amazon.de for €69 and on MediaMarkt for €72. On Amazon, contribution before ads is €18.20. The team targets three competitor ASINs with Sponsored Display because those competitors have high traffic. After ten days, the campaign spends €420 and generates €1,100 in attributed sales: 38.2% ACOS. Technically below a launch ceiling, but above the SKU’s comfortable profit threshold.
The FiveX-style review asks what shoppers saw on those competitor pages. Competitor A is €62 with 4.6 stars. Competitor B is €74 but includes two extra ports. Competitor C ships tomorrow while VoltCase ships in three days. The campaign is not failing because Sponsored Display is bad. It is failing because the offer is not strong enough on two of the three pages.
The fix is surgical. Keep competitor B because the price comparison is favourable. Pause competitor A and C. Move €250 of weekly budget to Amazon Sponsored Products for “usb c hub 4k hdmi” where intent is clearer, and reserve a smaller MediaMarkt Sponsored Product Ads test for the same SKU if MediaMarkt stock is deeper. FiveX’s cross-marketplace budget view helps make that call without turning it into Amazon-versus-MediaMarkt politics.
Named example 3: the bol refill that belongs in purchase remarketing logic
Sponsored Display is Amazon-specific, but the operating idea travels across marketplaces. Suppose a brand sells “FreshBrew Water Filters” on Amazon.nl and bol.com. The pack sells for €24.95, with €7.40 contribution before ads on Amazon and €6.80 on bol after LVB and marketplace costs. Repeat purchase usually happens after 42–55 days.
On Amazon, purchase remarketing with a 45-day lookback makes commercial sense. The campaign gets €450 for the month, a target ACOS ceiling of 24%, and a message focused on refill timing. On bol, the team cannot copy the exact Sponsored Display mechanic in the same way, but it can use Sponsored Products and CRM timing around the same replenishment window.
The lesson: do not evaluate Amazon retargeting in isolation. If Amazon purchase remarketing lifts repeat orders but bol stock is tighter and bol margin is lower, the next euro may still belong on Amazon. FiveX’s SKU mapping and multi-channel profitability view make that decision visible at product-family level instead of platform-report level.
Budget rules for €5K+ marketplace ad accounts
For an account spending €5K per month, I would not let Sponsored Display eat a heroic share of budget until it proves incremental value. A simple starting allocation looks like this:
- 60–70% Sponsored Products: search demand, ranking support, product targeting and harvesting.
- 15–25% Sponsored Brands and video: category visibility, Store traffic and brand-led discovery.
- 5–15% Sponsored Display: self-targeting, views remarketing and tightly filtered conquesting.
- 0–10% test budget: vCPM, broader audiences, DSP learning or cross-marketplace retail media tests.
The percentages are less important than the permission logic. Sponsored Display earns more budget when total SKU contribution rises, stock remains healthy, TACOS does not become more dependent on paid traffic, and the audience segment has a clear job. It loses budget when reported ROAS improves but product-family profit does not.
How to measure Sponsored Display without fooling yourself
Use three layers of measurement.
Campaign layer
Track spend, sales, ACOS, CTR, CPC, conversion rate, advertised product sales and purchased product sales. Split CPC and vCPM campaigns in reporting. Mixing them makes the meeting too optimistic.
SKU layer
Track contribution margin after ads, return rate, current price, Buy Box or Featured Offer status, stock cover and organic rank movement. A campaign that sells a SKU with weak retained margin is not “efficient”. It is merely well attributed.
Business layer
Track TACOS, total product-family sales, new-to-brand share where available, repeat purchase rate and on/off lift. If pausing a Sponsored Display campaign barely moves total sales, the campaign was probably harvesting credit rather than creating demand.
FiveX AI recommendations can help here by surfacing the awkward but useful moves: “pause retargeting because stock cover is 12 days”, “move budget from competitor ASINs to self-targeting”, or “lower bid because post-return margin is below threshold”. The best recommendation is often not more spend. It is a polite no with numbers.
The weekly operating rhythm
Run Sponsored Display in a weekly rhythm, not as a set-and-forget retargeting toy.
- Monday: refresh SKU margin, stock cover, price position and Buy Box status.
- Tuesday: review campaign role: defence, retargeting, conquesting, replenishment or reach.
- Wednesday: adjust bids and budgets only for SKUs with permission to scale.
- Thursday: check search and PDP overlap with Sponsored Products and Sponsored Brands.
- Friday: decide one action per campaign: scale, cap, narrow, test, pause or move budget to another marketplace.
This sounds operational because it is. Sponsored Display performance is not created by knowing where the buttons are. It is created by refusing to let one ad format blur five commercial jobs.
Final takeaway
Amazon Sponsored Display can absolutely deserve budget. It can defend product pages, recover warm shoppers, support repeat purchase, cross-sell accessories and create useful audience reach. But it is not automatically profitable because the audience is warm, and it is not automatically incremental because Amazon reports sales after exposure.
The operator’s rule is simple: scale Sponsored Display only when the SKU, stock and incrementality case all agree. If one of those three says no, keep the campaign small, narrow the audience or move the money back to search, bol or MediaMarkt where the next euro has a better job.
That is how Sponsored Display becomes a profit tool instead of a very elegant attribution trap.