I wouldn't force a return to China unless the item is expensive enough to justify it. On most orders, the return shipping and hassle cost more than the product, and pushing it often ends in a chargeback, where the processor usually sides with the buyer. How I'd decide: 1. Run the math: product cost, refund fees, and your time versus the chargeback risk and a bad review. If it's a cheap item, refund, let them keep it, and close it out. 2. For higher-priced items, offer one clear choice in writing: a full refund on return, or a partial refund to keep the item, with a reply-by date. 3. Note the reason. If the same complaint keeps coming up on this product, it's a supplier or product page problem, not a one-off customer. On a normal low to mid priced order, I'd refund and move on. Save the hard return rule for expensive products where you actually get something back.

@Artjom Milantejs my pleasure!

Jeremy's path is right. You can't really scale a product with no conversion data yet. You buy cheap learning first and only raise spend once it proves it can sell. 1. Put the no-data products in their own test campaign so they can't take spend from your winners. 2. Start on Max Clicks with a hard max CPC. The goal early on is cheap clicks, not ROAS. Smart bidding with almost no conversions tends to go cautious or random. 3. Fix the offer before adding budget. That means a clean title, a strong main image, correct price and availability, and a product page that matches the ad. A lot of "no data" products really have a page problem. 4. Give each product a kill rule, for example a set amount of spend or clicks with no add-to-carts, then pause. 5. Once a product converts, move it to smart bidding and raise budget in steps. If you share your daily budget, campaign type, and whether these products get impressions at all, people can tell you whether the problem is bids, feed, or offer.

Agree with Milica on Manual CPC. Brand clicks should be cheap, so start low and only raise bids if you're losing exact brand searches. Zero impressions for 3 days usually means something besides the bid strategy, though. Before switching, check: 1. Eligibility. Ads approved, brand keywords on exact or phrase match, and locations and languages matching where you sell. 2. PMax taking the brand traffic. If PMax is already winning your brand searches, the Search campaign looks dead even though the account is fine. Check search terms on both campaigns. This also ties into the brand negatives we talked about on Mikkel's thread. 3. Target Impression Share choking spend. A high target with a low max CPC cap will sit at zero. Brand Search rarely needs it unless competitors or affiliates are bidding on your name. Switch to Manual CPC, confirm the keywords and locations are clean, then decide whether you need a separate brand campaign at all.

Underspending for a couple of days right after a split is normal. You cut the main campaign down to about 20 products overnight, so it's matching fewer searches and Google needs a few days to relearn who to show. Before you change anything, check: 1. Are all 20 products in the main campaign still approved in Merchant Center? Disapprovals or stock issues after the move will cap spend fast. 2. Did impression share drop? The split removed traffic the broader feed used to get. That can be fine if what's left is cleaner, but spend will stay low until volume builds back. 3. If you're on a tight tROAS, fewer conversions can make Google bid cautiously. Give it a few days of clean data before loosening the target, and change one thing at a time. 4. The €20 non-sellers campaign may never fully spend. Treat it as a cheap filter, not a second growth engine. I'd wait a few more full days and confirm Merchant Center is clean before deciding anything. Don't undo the split because the first two days underspent.

More products on the same €300 a day usually doesn't create more winners. It spreads the budget across a bigger feed, so new SKUs and zombies fight your proven sellers for spend and never leave learning. What works better: 1. Protect your winners. Keep the main campaign on the converting products with enough budget to actually spend and convert. 2. Add new products in batches. Put 20 to 40 new SKUs in a separate test campaign with a fixed slice, around €40 to €60 of the €300. Give each batch enough time to show a signal, pause the duds, then add the next batch. 3. Fix the feed before adding more. New SKUs need clean titles, GTINs if you have them, strong main images, accurate price and availability, and a product page that matches. 4. Watch product-level data. Zero impressions after a week usually points to a feed or budget problem. Clicks but no sales points to the product page or the offer. With €300 a day you can hunt for new winners, but only if the hunt has its own budget and the winners keep theirs.