Walmart, eBay and Etsy Sponsored Placements on Margin

Ask for a target ACOS and the honest answer is that ACOS is the wrong denominator on a marketplace. On Walmart, a 15 percent referral fee against a 34 percent gross margin leaves nineteen points to spend, and an item running at 14 percent ad cost is earning five — before returns, which in home and furniture take three to six of those points back. I build every account on contribution after fee, returns and ad cost at item level, and roughly a quarter of the catalogue usually turns out to be worth no advertising at all. Walmart Connect: Automatic campaigns run first, purely to harvest search terms, then convert into manual exact with placement bid multipliers set separately for search in-grid, buy box and item page. Those three placements rarely deserve the same bid, and treating them as one is the most common thing I fix. Eligibility comes before bidding: an item without the Buy Box does not serve, and listing content quality gates both organic placement and ad relevance, so the first fortnight is usually attribute and image work on the leading items rather than campaign work. I report the three-day attribution window alongside the fourteen-day default, because the gap between them tells you whether you are buying decided purchases or creating them. eBay: Promoted Listings Standard charges an ad rate on the sale; Advanced charges per click. The judgement is which listings belong on which, and the split is not about revenue size. Slow-moving, high-margin inventory earns Standard, where you pay on the outcome and the trending rate is a ceiling to argue with rather than a rule to follow. Items with real search demand and a price you can defend earn Advanced, with keyword control and negatives. Running the whole catalogue on Standard at the suggested rate is a tax, not a strategy. Etsy: Etsy Ads exposes no keyword bidding and no bid control, so anyone selling you Etsy keyword optimisation is selling you tags. The levers that exist are which listings are eligible to spend, the daily budget, and the attributes and tags that decide which queries a listing can match at all. In practice the work is switching off the listings absorbing budget without orders, rewriting titles and attributes on the shop's strongest sellers, and setting the budget where one week produces enough orders to read. Not included: Product photography and video. Rewriting every listing — I rewrite the items carrying the ad spend and give your team the pattern for the rest. Amazon Ads: I do not run it, the specialist pool there is deep and I would be a mediocre choice. Inventory buying and repricing decisions. Marketplace account health disputes and suspension appeals. Who this is not for: Sellers who do not hold the Buy Box on their leading items, because advertising into a lost Buy Box funds the winner. Catalogues where price is set centrally and cannot move on any item. Anyone who wants one dashboard blending marketplace ads with Google and Meta into a single ROAS figure: the attribution models are not compatible and combining them produces a number nobody should act on. Reporting: Item-level contribution rather than campaign ROAS, refreshed weekly, with the items I want removed from advertising listed first and the reason next to each.