Christopher's right that a hard tROAS on ~1 order/day and $59 spend is usually too tight. The bigger issue is you changed three things at once, so you can't tell which one killed volume. I'd unwind in this order: 1) Revert Target ROAS back to Maximize Conversion Value (or a much looser target) and leave the product set and shipping offer alone for a few days. On thin data, a higher tROAS often just stops the auction rather than "optimizing." 2) If volume returns, reintroduce one change at a time. Dead-product cleanup is usually fine. Free shipping above a threshold can quietly wreck checkout if a lot of carts sit under the threshold, so watch CVR, not just ROAS. 3) Don't wait weeks hoping learning fixes a target that was set above what the account can sustain. Four to five quiet days with normal clicks and spend already says the bid constraint (or the offer change) choked conversion, not "needs more time." Stabilize on Max Conv Value first. Then test shipping or product cuts separately so the next drop has one cause.

Your follow-up is the smoking gun: people needed the item by a date, saw no shipping link after a day, and canceled. That's expectation, not "cheap buyers hate waiting." What I'd tighten: 1) Product page and ads: only promise what you can hit every time. "Within 48 hours" is safer than next-day if next-day isn't real. Match the same language in your Shopping feed shipping settings so Google isn't selling a faster story than the site. 2) Right after purchase: a confirmation email/SMS with the real timeline ("We ship within 48 hours; tracking lands once the label is created") and what "processing" looks like on day one. Silence after checkout is what triggers the cancel. 3) Cheap SKUs do raise urgency, but the fix is still honesty and updates, not higher prices. If a segment always needs it tomorrow, either offer a paid expedited option or stop buying that traffic with a speed claim. Christopher's right to ask customers. I'd also watch cancel rate by product, and any "needed by" mentions, for a week after the copy change. If cancels drop when the promise matches ops, you don't need a deeper product redesign.

Luis is right that you have to fix tracking. Scaling on a number Google can't see is how you teach the algo the wrong lesson. How I'd split the decisions while you repair it: 1) Use your own Shopify/UTM numbers for "is this profitable enough to raise budget?" That's the cash-flow truth. 2) Use Google's reported conversions for "is Smart Bidding healthy?" If Google only sees a fraction of orders, Max Conv Value / tROAS is optimizing on incomplete data. Raising budget on that signal often just spends more on whatever Google thinks is converting. 3) Yesterday's gap (0.59 in Ads vs ~2.67 on your side) is a tracking problem first, a scale problem second. Until primary conversions fire reliably (purchase tag plus enhanced conversions, or offline import if needed), treat Ads ROAS as a lower bound, not the decision number. Practical order: verify the purchase conversion fires on the thank-you page, check that enhanced conversions / consent mode aren't dropping half the events, and compare Google Ads vs Shopify order counts over 7 days. Once the gap shrinks, scale off Google's number. Until then, only small budget bumps if your own attributed profit is clearly above break-even, and don't tighten a tROAS target on undercounted data.

At €70/day with only 8–10 of 80 products selling, the structure problem is usually mixed signals, not "too few campaigns." I'd do this before raising budget: 1) Split winners from the rest. Put the 8–10 proven sellers in their own Shopping/PMax with the budget you actually want to protect. Leave the other ~70 in a separate low-budget test campaign (or pause the real zombies). One campaign feeding both proven and dead SKUs makes Smart Bidding chase noise and explains wild 0.8 to 10+ days. 2) Scale on cash-flow math, not weekly average ROAS. If ROAS swings that hard day to day, raise by ~20% (toward €85–100), then leave it several days. Jumping straight to €100 on a thin account often just buys more of the bad days. 3) Judge scale readiness on a 7-day blended ROAS after refunds/cancellations, plus whether the winners still spend when you give them their own budget. If the winner campaign eats the increase cleanly and ROAS holds near your break-even, keep going. If spend piles onto weak SKUs, the split was the fix, not more budget. Don't optimize the 70 zombies inside the same campaign as the money makers. Separate them, then scale the winners.

Seven days under review with Next greyed out usually isn't Google forgetting you. It often won't let you escalate until the account basics are green. Checklist I'd run: 1. Merchant Center is linked to the right Google Ads account and the link is approved, not linked to a second unused Ads account. 2. Billing is active on that Ads account. Without it, reviews can sit forever. 3. Open the products in Diagnostics and confirm whether they're really under review or disapproved for a reason like GTIN, image, landing page, or misrepresentation. 4. The product URL loads, ships to the target country, price and availability match the feed, and the item is allowed for Shopping. 5. First approvals on a new account can take several days even when everything is clean. After a week with Next still grey, something upstream is usually blocking it: the link, verification, or product eligibility. Fix the link and diagnostics first. If the products are clean and still pending, contact support with the offer IDs and the date they went under review.