Amazon DSP Behind a Retail Readiness Gate

We turn down a meaningful share of Amazon DSP briefs at the readiness stage, before quoting. Display against an ASIN holding the Buy Box seventy percent of the time, or one that goes out of stock for nine days a quarter, converts badly for reasons no bid strategy repairs. The gate is the same every time: Buy Box share above ninety percent across the trailing thirty days, four or more weeks of cover, at least fifteen reviews, and a rating above 3.7. Fail two of those and we say no and tell you what to fix. The readiness gate: It is run on your Seller Central or Vendor Central data, not on scraped estimates. Out-of-stock history matters more than the rating; a chilled or short-life product with erratic replenishment will burn budget on detail pages that cannot fulfil. For food and drink specifically we look at case pack economics before anything else, because a three-euro unit sold as a single with prepaid returns loses money at any conversion rate display could plausibly produce. Where Irish brands actually sell: Irish demand largely lands on the UK and German storefronts rather than a domestic one, which changes the buy. It means UK VAT registration and EORI arrangements are a precondition rather than a detail, that reviews accumulate on a marketplace where your brand has no shelf recognition, and that the media plan should be written against the UK catalogue you can actually keep in stock, not the full range you sell at home. Amazon Marketing Cloud: Attribution inside the DSP interface counts view-through generously. We work in AMC instead, writing clean-room SQL for path-to-conversion, new-to-brand share and overlap between Sponsored ads and DSP. The measurement that decides renewal is a holdout built in AMC: a matched audience deliberately left unexposed, read at ninety days. It reliably reports a smaller number than the platform does. That smaller number is the one we plan the next quarter on. Inventory: Amazon-owned placements first, meaning detail page, search results shelf, Fire TV, Twitch and Freevee where the audience fits. The third-party display tail bought through Amazon's supply gets used sparingly and only with a named deal behind it, because that inventory is where DSP spend quietly turns into low-quality reach that still reports conversions through view-through. Overlap with Sponsored ads: We do not take over Sponsored Products or Sponsored Brands. We do read the Search Query Performance report, because a DSP plan built without knowing which queries already convert is guesswork. If your Sponsored campaigns are not yet saturating the profitable queries, DSP is the wrong next euro and we will say so at the gate. Not included: Listing copy, photography, A+ content, Brand Store build, Vendor terms negotiation, and marketplace account health work. We will tell you which of those is blocking the buy. We will not do them. Walk away if: You are pre-launch on Amazon with no review base. You want display to compensate for a listing problem. Your annual marketplace revenue is small enough that the DSP minimums and this fee together exceed what Sponsored Products would return on the same money. Or the deciding metric is the DSP's own attributed return, in which case someone else will be happy to report it to you.