Efficiency metrics (ACOS −0.5pp, ROAS +1.6%) improved while every volume metric fell: orders 557 → 495, CVR 3.28% → 2.95%, ad sales $47,081 → $45,479. Spend was nearly flat (−4.9%), so this is not an efficiency problem — the account is converting fewer of the same clicks and is throttled on the volume side.

The window does not overlap Prime Day or a major holiday, so the read is clean. Note Amazon's 2–14 day attribution lag: the final days of the window may revise upward.

B0SAMPLE01 takes 5.4% of spend ($776) but returns 0.9% of ad sales — a 98.4% ACOS that single-handedly drags the account. Either it needs a dedicated low-bid defensive structure or it should be paused pending a listing/price review.

The two efficient hero ASINs (B0SAMPLE02, B0SAMPLE03) are slightly under-spent relative to the sales they generate — room to fund them from the sink above.

SP dominance is correct for a mature tools catalog. The only gap is Sponsored Brands at 4.1% — slightly under the 5–15% awareness range; one branded headline campaign on the hero ASINs would close it. No dynamic-bidding "down only → up & down" risk flags on the top campaigns. Naming is consistent with the VT convention.

Set placement modifiers to track relative CVR, not ACOS. Top of Search converts at 2.8× Product Pages, so it warrants the largest bid uplift. Across the account, modifiers already track this — except campaign SP-Auto-Sample-US, which sets Top of Search at +12% while Product Pages sits at +50%, exactly inverted. Recommend ToS +90% / PP 0% on that campaign.

Recommended bid for non-converters = AOV × target ACOS ÷ clicks-to-conversion, floored at the placement's break-even. High-ACOS targets are capped toward target ACOS using the account's actual RPC, not a flat haircut. Benchmarks (AOV $91.80, aCTC 34 clicks, CVR 2.95%) are derived from this profile's aggregates — not hardcoded.

Harvest = orders > 1 AND CVR ≥ the 2.95% account benchmark AND clicks ≥ aCTC (34), currently running in Auto or Broad. Negate = orders = 0 with clicks ≥ aCTC. Promote harvest winners to Exact in the matching campaign; add the waste terms as campaign or ad-group negatives so the low-intent cluster stops recurring.

Over a quarter of spend is in Auto — fine for a young account, too high for a mature catalog. The winners Auto has already surfaced (see §06 harvest) should graduate to Exact Non-Brand, shifting budget from undirected discovery (26.7% → ~15%) into the Non-Brand core (51% → ~73%). Competitor and Brand defense are healthy.

The Exact gap and the harvest backlog are the same story: proven terms are stuck in Broad/Auto. Separately, 9 high-performing Exact campaigns are budget-capped (hit budget before day's end >20 days this window) at a blended 22% ACOS — open these budgets to recover throttled volume. Full list in the account export.